Patent Application
20240320923
Designing a system to safely drive a vehicle autonomously without supervision is tremendously difficult. An autonomous vehicle should at least be capable of performing as a functional equivalent of an attentive driver who draws upon a perception and action system that has an incredible ability to identify and react to moving and static obstacles in a complex environment-to avoid colliding with other objects or structures along the path of the vehicle. Thus, the ability to detect animate objects (e.g., cars, pedestrians, etc.) and other parts of an environment is often critical for autonomous driving perception systems. Many conventional perception methods rely on deep neural networks (DNNs) to evaluate images or other sensor data for detection tasks.
One of the most significant challenges in visual perception for autonomous driving arises from the diversity in camera rig configurations (e.g., different number of cameras, fields of view, positions, orientations, etc.) implemented by different vehicle manufacturers across different vehicle models. Unlike two-dimensional (2D) computer vision, where DNNs may offer some translation invariance and may therefore be somewhat robust to certain changes in camera position, multi-view or three-dimensional (3D) perception (e.g., that relies on lifting 2D (e.g., image) features into 3D space) often experiences degraded performance when processing data that was generated using different camera rigs, especially when the road is nonplanar. As a result, the diversity of camera rig configurations across different vehicle models (e.g., sedans, trucks, sport utility vehicles) limits the ability to develop generalized DNNs that perform well across different vehicle models. Ideally, DNNs would be trained using data generated by data collection vehicles equipped with the same camera rig as the target vehicle model (e.g., same number of cameras, viewpoint, camera parameters) to minimize the domain gap. However, this is often unfeasible in the real world. For example, collecting new data and developing a dedicated DNN for each vehicle model is highly resource intensive and impractical, and is often impossible for new vehicle models. As such, there is a need for improved perception techniques that can support multi-view or 3D perception.
Embodiments of the present disclosure relate to viewpoint adapted perception for autonomous machines and applications. For example, systems and methods are disclosed that adapt a 3D perception network to accommodate unavailable target rig data. In some embodiments, a 3D perception network may be adapted to accommodate unavailable target rig data by training one or more layers of the 3D perception network as part of a training network using real source rig data and simulated source and target rig data. Feature statistics extracted from the real source data may be used to transform the features extracted from the simulated data during training. The paths for real and simulated data through the training network may be alternately trained on real and simulated data to update shared weights for the different paths.
Additionally or alternatively, the 3D perception network may be adapted to accommodate unavailable target rig data by training one or more layers of the 3D perception network as part of a training network using simulated source and target rig data without real source rig data. A consistency loss that compares (e.g., top-down) transformed feature maps extracted from simulated source and target rig data may be used to minimize differences across training channels.
As such, one or more of the paths through the training network(s) may be designated as the 3D perception network, and target rig data may be applied to the 3D perception network to perform one or more perception tasks.
The present systems and methods for viewpoint adapted perception for autonomous machines and applications are described in detail below with reference to the attached drawing figures, wherein:
embodiments of the present disclosure;
Systems and methods are disclosed related to viewpoint-adaptive perception for autonomous machines and applications. For example, to adapt a 3D perception network to accommodate unavailable target rig data, inter-domain or coarse viewpoint adaptation may be performed by training the 3D perception network using real source rig data and simulated source and target rig data, and/or inter-domain or coarse viewpoint adaptation may be performed by training the 3D perception network using simulated source and target rig data. As such, the 3D perception network may be trained to perform (e.g., top-down) object detection to identify or detect objects (e.g., cars, trucks, pedestrians, cyclists, etc.) and/or portions of the environment from (e.g., perspective view) target rig data. The present techniques may be used to generate a viewpoint-adaptive or rig-invariant DNN for autonomous vehicles, semi-autonomous vehicles, robots, and/or other object types.
Although the present disclosure may be described with respect to an example autonomous or semi-autonomous vehicle or machine 700 (alternatively referred to herein as “vehicle 700” or “ego-machine 700,” an example of which is described with respect to
At a high level, a DNN may be used to perform some type of 3D perception using 2D sensor data representing a 3D environment, and may rely on 3D reconstruction techniques such as a 2D-to-3D transformation or 2D-3D fusion to extract or reason about 3D information from one or more 2D inputs. Some example tasks involving 2D-to-3D transformations include depth estimation from monocular images in which the DNN predicts 3D depth values from a 2D image, and view synthesis or multi-view reconstruction in which one or more 2D images are transformed into a 3D representation and/or some other 2D view (e.g., leveraging implicit 3D spatial relationships or representations). For example, the DNN may include multiple constituent DNNs or stages chained together, sequentially processing different views of the 3D environment. An example multi-view 3D perception DNN may include an input stage that extracts features in a first view (e.g., perspective view) or a plurality of first views (e.g., applying images generated from different perspectives to different input channels), a transformation to a second view (e.g., top-down or bird's eye view), and an output stage that operates (e.g., performs class segmentation and/or instance regression) in the second view. Some example tasks that integrate or fuse information from both 2D and 3D sources include pose estimation in which the DNN uses 2D projections of observations of objects to estimate their 3D pose, and augmented reality applications in which the DNN combines real-world 2D images or videos with virtual 3D objects for tasks such as object detection and tracking, pose estimation, semantic segmentation, depth estimation, or image-based rendering. The applicable sensor (e.g., camera) rig will typically depend on the specific 3D reconstruction task.
In some embodiments, the DNN may be trained to process target rig data using real source rig data and simulated source and target rig data. For example, a feature extractor of the DNN may be used as a shared feature extractor shared across different channels or branches, or replicated into multiple channels of a shared feature extractor, comprising a first channel(s) for real source rig data (e.g., images generated using an available camera rig configuration), a second channel(s) for simulated source rig data (e.g., simulated from the perspective of a virtual camera rig configuration corresponding to the available camera rig configuration), and a third channel(s) for simulated target rig data (e.g., simulated from the perspective of the target rig configuration). The output of each channel may be transformed from a first (e.g., perspective) 2D view to a second (e.g., top-down) 2D view and applied to a corresponding output (e.g., classification) channel. To address domain gap challenges, during training, feature statistics extracted from the real source data may be used to transform the features extracted from the simulated data. The paths for real and simulated data through the resulting training network may be alternately trained on real and simulated data to update shared weights for the different paths. This type of training may be considered coarse viewpoint adaptation as the updates may effectively train (update) the DNN to handle (e.g., process, reconcile, etc.) relatively large differences between source and target viewpoints and corresponding rig configurations (e.g., different number of cameras, fields of view, relatively larger changes in positions or orientations, etc.).
Additionally or alternatively, the DNN may be trained (updated) to process target rig data using simulated source and target rig data. For example, a feature extractor of the DNN may be replicated into multiple channels of a feature extractor comprising a first channel(s) for simulated source rig data (e.g., simulated from the perspective of a virtual camera rig configuration corresponding to the available camera rig configuration), and a second channel(s) for simulated target rig data (e.g., simulated from the perspective of the target rig configuration). The output of the first channel(s) may be transformed from a first (e.g., perspective) 2D view to a second (e.g., top-down) 2D view, applied to an output (e.g., classification) channel(s), and used to compute a corresponding (e.g., cross-entropy) loss. The output of the second channel(s) may be transformed from the first (e.g., perspective) 2D view to the second (e.g., top-down) 2D view, compared to the transformed output from the first channel(s), and used to compute a (e.g., top-down or bird's eye view) consistency loss that minimizes differences between transformed features across the channels. As such, the resulting training network may be trained using pairwise simulated data representing the same scene content simulated with different rig configurations, with the consistency loss effectively enforcing viewpoint invariance in the viewpoint transformation. This type of training may be considered fine viewpoint adaptation as the updates may effectively train the DNN to handle relatively smaller differences between source and target viewpoints and corresponding rig configurations (e.g., relatively smaller changes in camera positions or orientations).
In some embodiments, simulated rig data may be generated using a simulation system (e.g., NVIDIA DRIVE Sim™, Car Learning to Act (CARLA), Grand Theft Auto 5) and/or a simulated dataset (e.g., SYNTHetic collection of Imagery and Annotations (SYNTHIA), Virtual Karlsruhe Institute of Technology and Toyota Technological Institute (Virtual KITTI)). Taking DRIVE Sim as an example, a virtual source camera rig representing an available source camera rig (e.g., comprising six cameras) and a virtual target camera rig representing a target camera rig (e.g., comprising five or six cameras) may be generated in DRIVE Sim and used to generate a set of simulated data (e.g., images from the perspective of each of the cameras, corresponding ground truth data such as 2D/3D bounding boxes, a depth map, optical flows, a segmentation mask, etc.) for each of a plurality of simulated scenes and/or time slices. As such, source and target camera sensor rigs may be simulated and used to generate simulated training data, and the simulated training data may be used to adapt a DNN from a source viewpoint to a target viewpoint.
As such, the techniques described herein may be used to adapt a DNN to detect and classify animate objects and/or parts of an environment from (e.g., perspective view image) data generated using a target rig that differs from the source rig used to generate the training data, and these detections and classifications may be provided to an autonomous vehicle drive stack to enable safe planning and control of the autonomous vehicle. Unlike conventional approaches, the present techniques may be used to train or update a 3D perception DNN that performs well across different rig configurations and vehicle models, even when data from a target rig configuration is unavailable. Furthermore, the present techniques avoid the highly resource intensive process of collecting new data and developing a dedicated DNN for each vehicle model. As such, a DNN trained using the present techniques should yield improved detection accuracy on unavailable target rig configurations and substantially reduce the use of computational resources over prior techniques.
With reference to
As a high level overview, the viewpoint adaptation system 100 may adapt a 3D perception network to handle unavailable target rig data. For example, a neural network development component 101 may accept a designation of, construct, or otherwise identify a coarse viewpoint adaptation network 125 and/or a fine viewpoint adaptation network 150 comprising one or more layers of the 3D perception network to be viewpoint adapted. A training data generator 105 may generate or otherwise identify training data 175 for the coarse viewpoint adaptation network 125 and/or the fine viewpoint adaptation network 150, and a training component 110 may train or update the coarse viewpoint adaptation network 125 and/or the fine viewpoint adaptation network 150 using the training data 175. After training, the neural network development component 101 and/or the training component 110 may designate the one or more trained layers as the viewpoint adapted network 170.
Generally, the viewpoint adaptation system 100 and its component(s) may be implemented using a deep learning platform such as TensorFlow, PyTorch, Keras, MXNet, Caffe, Microsoft Cognitive Toolkit (CNTK), Fast.ai, or Theano. As such, the viewpoint adaptation system 100 may include a neural network development component 101 of a deep learning platform that provides application programming interfaces (APIs) and libraries to define and create neural network architectures. For example, the neural network development component 101 may accept input designating the number and configuration of network layers (e.g., convolutional, recurrent, etc.), activation functions, initialized network weights, etc. Further, the viewpoint adaptation system 100 may include a training component 110 of a deep learning platform that provides configurable training algorithms and optimization techniques. For example, the training component 110 may accept input designating loss functions, learning rate, batch size, optimization algorithms (e.g., stochastic gradient descent, Adam optimizer), regularization techniques, and other parameters or algorithms to customize a desired training process.
For example, the neural network development component 101 and/or the training component 110 may be used to create a viewpoint adapted network 170. More specifically, the neural network development component 101 may accept a designation of, construct, or otherwise identify a coarse viewpoint adaptation network 125 and/or a fine viewpoint adaptation network 150 using one or more layers that, once trained by the training component 110, may be used as at least a portion of the viewpoint adapted network 170.
More specifically, the neural network development component 101 may accept a designation of, construct, or otherwise identify the coarse viewpoint adaptation network 125, the fine viewpoint adaptation network 150, and/or the viewpoint adapted network 170 as a 3D perception network that accepts or otherwise processes a representation of 2D sensor data (e.g., image data) that represents a 3D environment. For example, the 3D perception network may accept and process a representation of sensor data generated or simulated using a real or simulated sensor rig of an ego-machine, such as the example autonomous vehicle 700 of
Depending on the task and/or embodiment, the 3D perception network may use any known architecture and may include any number of input paths (e.g., heads, channels, layers, branches, etc.), transformations, fusion operations, and/or output paths (e.g., heads, channels, layers, branches, etc.). Taking an autonomous machine application as an example, the 3D perception network may include an input path comprising one or more layers (e.g., of a feature extractor) for each of one or more sensors (e.g., cameras) of an autonomous machine (e.g., the example autonomous vehicle 700 of
Continuing with the example multi-view 3D perception network, one or more viewpoints of the multi-view 3D perception network may be adapted to accommodate one or more target viewpoints represented by a target sensor rig by training (or finetuning, updating, etc.) one or more of its layers (e.g., the feature extractor(s), view transformer, and/or classifier(s)) as part of the coarse viewpoint adaptation network 125. In the example illustrated in
For example, the neural network development component 101 of
To address domain gap challenges, a style injector 235 (e.g., corresponding to the style injector 135 of C×H×W, the style injector 235 may perform a transformation t (
) such as:
where μ, σ∈C are the mean and standard deviation of the respective feature computed across spatial dimensions:
and where C, H, and W indicate channel dimension, height, and width. The superscripts s and r represent simulated and real, respectively. Fchw represents an element at position (c, h, w) of F, and ε may be set to 1×10−5. For example, a feature extractor may include any number of layers that generate any number of feature maps, and the style injector 235 may compute statistics from the feature maps extracted by any of the layers in the path for real data and inject the computed statistics into corresponding feature maps extracted by corresponding layers in the path for simulated data. This style injection may serve to minimize the style and content gap between extracted features from the real and simulated paths.
The training component 110 of
where Ŷ represents a predicted segmentation map, Y represents a ground truth segmentation map, and the superscripts rs, ss, and st stand for real source, simulated source, and simulated target, respectively. As such, the coarse viewpoint adaptation component 115 may use equation 4 to compute a cross-entropy loss ce based on the real source rig data 210, may use equation 5 to compute a cross-entropy loss
ce based on the simulated source rig data 215 and the simulated target rig data 220, and may use equation 6 to compute a loss
that combines the losses based on real and simulated data. The balancing hyper-parameter λaux may be set to 1. As such, the coarse viewpoint adaptation component 115 may alternate applying the real source rig data 210 on the one hand and the simulated source and target rig data 215 and 220 on the other to the coarse viewpoint adaptation network 200, calculate a combined loss, and update the coarse viewpoint adaptation network 200 accordingly.
Additionally or alternatively, and continuing with the example multi-view 3D perception network, the multi-view 3D perception network may be viewpoint adapted to handle one or more target viewpoints represented by a target sensor rig by training one or more of its layers (e.g., the feature extractor(s), view transformer, and/or classifier(s)) as part of the fine viewpoint adaptation network 150. In the example illustrated in
For example, the neural network development component 101 of
The training component 110 of
where Xsrc, Xtgt∈C×L×W are the transformed (e.g., top down or bird's eye view) feature maps extracted from the simulated source rig data 315 and the simulated target rig data 320, and C, L, and W represent the number of channels, length, and width of the feature maps, respectively. In an example implementation involving segmentation, the following training objective may be used:
where λbev is a balancing hyper-parameter that may be selected by design. As such, the fine viewpoint adaptation component 120 may train the fine viewpoint adaptation network 200 using pairwise simulated data representing the same scene content simulated with different rig configurations, with the consistency loss effectively enforcing viewpoint invariance in the viewpoint transformation.
As such, and returning to
In some embodiments and returning to
For example, real source rig data 180 may be collected using one or more data collection machines (e.g., data collection vehicles), such as the example autonomous vehicle 700 of
As such, the training data generator 105 may access the real source rig data 180 (e.g., frames of image data generated using one or more data collection vehicles), and may generate or receive corresponding ground truth training data (e.g., classification data such as a top-down segmentation map) that matches the size and dimensionality of the corresponding model inputs and outputs. As such, the training data generator 105 may associate corresponding real frames of input training data (e.g., sets of frames of real image data generated using a source sensor rig) and ground truth training data (e.g., generated or annotated classification data such as a top-down segmentation map).
In some embodiments, the training data generator 105 includes or interfaces with a
simulation system that generates simulated source rig data 185 and/or simulated target rig data 190. The type of simulation system may depend on the use case or application. For example, the training data generator 105 may include or use a physics-based simulation engine like Unreal Engine or Unity3D to create realistic virtual environments where simulated objects can interact. Depending on the use case and desired training data 175, a physics-based simulation engine may be used to simulate various scenarios such as indoor or outdoor scenes and environments. In some embodiments, the training data generator 105 may include or use a domain-specific simulation platform. Depending on the implementation, the training data generator 105 may include or interface with a simulation system that simulates the training data 175 in any suitable simulation domain, such as simulations of vehicle environments, robotics environments, augmented and virtual reality (AR/VR) environments, industrial or manufacturing environments, medical or surgical environments, gaming environments, and/or others. Taking vehicle simulation as an example, the training data generator 105 may include or use a domain-specific simulation platform like NVIDIA Drive Sim or CARLA, which are specifically designed for autonomous driving simulations. These platforms offer detailed 3D environments, realistic vehicle models, and tools for generating diverse scenarios with different weather conditions, traffic patterns, and road layouts. In some embodiments, data augmentation may be applied within such a simulation system to create variations in lighting, textures, object positions, and/or other factors.
Taking DRIVE Sim as an example, DRIVE Sim may be used to generate physically accurate simulated sensor data and ground truth labels to facilitate the training of autonomous vehicle perception models. Simulation systems like DRIVE Sim may support real-time rendering, hardware-in-the-loop (HIL) simulation, and a high level of photo-realism. In autonomous driving, the value of simulated data (also known as synthetic data) derives in large part from difficulties in collecting and annotating real-world data, especially in the context of rare and critical events like uncommon objects, far-field objects, fatal accidents, and unusual driving scenarios. In situations involving an unavailable target sensor rig, the training data generator 105 may use a simulation system to generate simulated target rig data 190 representative of the target sensor rig.
Continuing with DRIVE Sim as an example, DRIVE Sim may ingest a representation (e.g., a Universal Scene Description (USD)) of a designated simulation comprising a scenario script, source and target sensor rigs, and a specified set of 3D assets (e.g., a model of the source and/or target vehicle model). For example, a scenario script may be authored in high-level scenario description language (HSDL), may define designated behaviors, events, and interactions within the simulated environment, and may be converted to a compatible format (e.g., USD). The script may outline domain-specific simulation scenarios such as traffic patterns, pedestrian movements, weather conditions, road layouts, vehicle behaviors, domain randomization, and/or otherwise. As such, the training data generator 105 may run the designated simulation in the simulation system and render a set of scenes, generating a frame of simulated sensor data (e.g., the simulated source rig data 185, the simulated target rig data 190) from the perspective of each simulated sensor (e.g., including lens distortion) and designated ground truth data (e.g., 2D/3D bounding boxes, a depth map, optical flows, a segmentation mask) corresponding to the output of the coarse viewpoint adaptation network 125, the fine viewpoint adaptation network 150, and/or the viewpoint adapted network 170.
For example, taking an example source sensor rig comprising six cameras, an example target sensor rig comprising six cameras, and an example 3D perception network (e.g., the coarse viewpoint adaptation network 125, the fine viewpoint adaptation network 150, the viewpoint adapted network 170) that performs top-down class segmentation in its output stage, the training data generator 105 may generate training data 175 comprising simulated input data (e.g., simulated images for each of the six cameras) and simulated ground truth data (e.g., labeled 3D bounding boxes, a labeled top-view segmentation mask, etc.) for each sensor rig and each time slice of a plurality of time slices representing different simulated 3D scenes. As such, the training data generator 105 may associate corresponding simulated frames of input training data (e.g., sets of frames of simulated image data simulated from the perspective of a source and/or target sensor rig) and ground truth training data (e.g., generated or annotated classification data such as a top-down segmentation map).
Now referring to
The systems and methods described herein may be used by, without limitation, non-autonomous vehicles, semi-autonomous vehicles (e.g., in one or more adaptive driver assistance systems (ADAS)), piloted and un-piloted robots or robotic platforms, warehouse vehicles, off-road vehicles, vehicles coupled to one or more trailers, flying vessels, boats, shuttles, emergency response vehicles, motorcycles, electric or motorized bicycles, aircraft, construction vehicles, trains, underwater craft, remotely operated vehicles such as drones, and/or other vehicle types. Further, the systems and methods described herein may be used for a variety of purposes, by way of example and without limitation, for machine control, machine locomotion, machine driving, synthetic data generation, model training, perception, augmented reality, virtual reality, mixed reality, robotics, security and surveillance, simulation and digital twinning, autonomous or semi-autonomous machine applications, deep learning, environment simulation, object or actor simulation and/or digital twinning, data center processing, conversational AI, light transport simulation (e.g., ray-tracing, path tracing, etc.), collaborative content creation for 3D assets, cloud computing, generative AI, and/or any other suitable applications.
Disclosed embodiments may be comprised in a variety of different systems such as automotive systems (e.g., a control system for an autonomous or semi-autonomous machine, a perception system for an autonomous or semi-autonomous machine), systems implemented using a robot, aerial systems, medial systems, boating systems, smart area monitoring systems, systems for performing deep learning operations, systems for performing simulation operations, systems for performing digital twin operations, systems implemented using an edge device, systems incorporating one or more virtual machines (VMs), systems for performing synthetic data generation operations, systems implemented at least partially in a data center, systems for performing conversational AI operations, systems implementing one or more language models-such as one or more large language models (LLMs), systems for performing light transport simulation, systems for performing collaborative content creation for 3D assets, systems implemented at least partially using cloud computing resources, and/or other types of systems.
The vehicle 700 may include components such as a chassis, a vehicle body, wheels (e.g., 2, 4, 6, 8, 18, etc.), tires, axles, and other components of a vehicle. The vehicle 700 may include a propulsion system 750, such as an internal combustion engine, hybrid electric power plant, an all-electric engine, and/or another propulsion system type. The propulsion system 750 may be connected to a drive train of the vehicle 700, which may include a transmission, to allow the propulsion of the vehicle 700. The propulsion system 750 may be controlled in response to receiving signals from the throttle/accelerator 752.
A steering system 754, which may include a steering wheel, may be used to steer the vehicle 700 (e.g., along a desired path or route) when the propulsion system 750 is operating (e.g., when the vehicle is in motion). The steering system 754 may receive signals from a steering actuator 756. The steering wheel may be optional for full automation (Level 5) functionality.
The brake sensor system 746 may be used to operate the vehicle brakes in response to receiving signals from the brake actuators 748 and/or brake sensors.
Controller(s) 736, which may include one or more system on chips (SoCs) 704 (
The controller(s) 736 may provide the signals for controlling one or more components and/or systems of the vehicle 700 in response to sensor data received from one or more sensors (e.g., sensor inputs). The sensor data may be received from, for example and without limitation, global navigation satellite systems (“GNSS”) sensor(s) 758 (e.g., Global Positioning System sensor(s)), RADAR sensor(s) 760, ultrasonic sensor(s) 762, LiDAR sensor(s) 764, inertial measurement unit (IMU) sensor(s) 766 (e.g., accelerometer(s), gyroscope(s), magnetic compass(es), magnetometer(s), etc.), microphone(s) 796, stereo camera(s) 768, wide-view camera(s) 770 (e.g., fisheye cameras), infrared camera(s) 772, surround camera(s) 774 (e.g., 360 degree cameras), long-range and/or mid-range camera(s) 798, speed sensor(s) 744 (e.g., for measuring the speed of the vehicle 700), vibration sensor(s) 742, steering sensor(s) 740, brake sensor(s) (e.g., as part of the brake sensor system 746), one or more occupant monitoring system (OMS) sensor(s) 701 (e.g., one or more interior cameras), and/or other sensor types.
One or more of the controller(s) 736 may receive inputs (e.g., represented by input data) from an instrument cluster 732 of the vehicle 700 and provide outputs (e.g., represented by output data, display data, etc.) via a human-machine interface (HMI) display 734, an audible annunciator, a loudspeaker, and/or via other components of the vehicle 700. The outputs may include information such as vehicle velocity, speed, time, map data (e.g., the High Definition (“HD”) map 722 of
The vehicle 700 further includes a network interface 724 which may use one or more wireless antenna(s) 726 and/or modem(s) to communicate over one or more networks. For example, the network interface 724 may be capable of communication over Long-Term Evolution (“LTE”), Wideband Code Division Multiple Access (“WCDMA”), Universal Mobile Telecommunications System (“UMTS”), Global System for Mobile communication (“GSM”), IMT-CDMA Multi-Carrier (“CDMA2000”), etc. The wireless antenna(s) 726 may also allow communication between objects in the environment (e.g., vehicles, mobile devices, etc.), using local area network(s), such as Bluetooth, Bluetooth Low Energy (“LE”), Z-Wave, ZigBee, etc., and/or low power wide-area network(s) (“LPWANs”), such as LoRaWAN, SigFox, etc.
The camera types for the cameras may include, but are not limited to, digital cameras that may be adapted for use with the components and/or systems of the vehicle 700. The camera(s) may operate at automotive safety integrity level (ASIL) B and/or at another ASIL. The camera types may be capable of any image capture rate, such as 60 frames per second (fps), 120 fps, 240 fps, etc., depending on the embodiment. The cameras may be capable of using rolling shutters, global shutters, another type of shutter, or a combination thereof. In some examples, the color filter array may include a red clear clear clear (RCCC) color filter array, a red clear clear blue (RCCB) color filter array, a red blue green clear (RBGC) color filter array, a Foveon X3 color filter array, a Bayer sensors (RGGB) color filter array, a monochrome sensor color filter array, and/or another type of color filter array. In some embodiments, clear pixel cameras, such as cameras with an RCCC, an RCCB, and/or an RBGC color filter array, may be used in an effort to increase light sensitivity.
In some examples, one or more of the camera(s) may be used to perform advanced driver assistance systems (ADAS) functions (e.g., as part of a redundant or fail-safe design). For example, a Multi-Function Mono Camera may be installed to provide functions including lane departure warning, traffic sign assist and intelligent headlamp control. One or more of the camera(s) (e.g., all of the cameras) may record and provide image data (e.g., video) simultaneously.
One or more of the cameras may be mounted in a mounting assembly, such as a custom designed (three dimensional (“3D”) printed) assembly, in order to cut out stray light and reflections from within the car (e.g., reflections from the dashboard reflected in the windshield mirrors) which may interfere with the camera's image data capture abilities. With reference to wing-mirror mounting assemblies, the wing-mirror assemblies may be custom 3D printed so that the camera mounting plate matches the shape of the wing-mirror. In some examples, the camera(s) may be integrated into the wing-mirror. For side-view cameras, the camera(s) may also be integrated within the four pillars at each corner of the cabin.
Cameras with a field of view that include portions of the environment in front of the vehicle 700 (e.g., front-facing cameras) may be used for surround view, to help identify forward facing paths and obstacles, as well aid in, with the help of one or more controllers 736 and/or control SoCs, providing information critical to generating an occupancy grid and/or determining the preferred vehicle paths. Front-facing cameras may be used to perform many of the same ADAS functions as LiDAR, including emergency braking, pedestrian detection, and collision avoidance. Front-facing cameras may also be used for ADAS functions and systems including Lane Departure Warnings (“LDW”), Autonomous Cruise Control (“ACC”), and/or other functions such as traffic sign recognition.
A variety of cameras may be used in a front-facing configuration, including, for example, a monocular camera platform that includes a complementary metal oxide semiconductor (“CMOS”) color imager. Another example may be a wide-view camera(s) 770 that may be used to perceive objects coming into view from the periphery (e.g., pedestrians, crossing traffic or bicycles). Although only one wide-view camera is illustrated in
Any number of stereo cameras 768 may also be included in a front-facing configuration. In at least one embodiment, one or more of stereo camera(s) 768 may include an integrated control unit comprising a scalable processing unit, which may provide a programmable logic (“FPGA”) and a multi-core micro-processor with an integrated Controller Area Network (“CAN”) or Ethernet interface on a single chip. Such a unit may be used to generate a 3D map of the vehicle's environment, including a distance estimate for all the points in the image. An alternative stereo camera(s) 768 may include a compact stereo vision sensor(s) that may include two camera lenses (one each on the left and right) and an image processing chip that may measure the distance from the vehicle to the target object and use the generated information (e.g., metadata) to activate the autonomous emergency braking and lane departure warning functions. Other types of stereo camera(s) 768 may be used in addition to, or alternatively from, those described herein.
Cameras with a field of view that include portions of the environment to the side of the vehicle 700 (e.g., side-view cameras) may be used for surround view, providing information used to create and update the occupancy grid, as well as to generate side impact collision warnings. For example, surround camera(s) 774 (e.g., four surround cameras 774 as illustrated in
Cameras with a field of view that include portions of the environment to the rear of the vehicle 700 (e.g., rear-view cameras) may be used for park assistance, surround view, rear collision warnings, and creating and updating the occupancy grid. A wide variety of cameras may be used including, but not limited to, cameras that are also suitable as a front-facing camera(s) (e.g., long-range and/or mid-range camera(s) 798, stereo camera(s) 768), infrared camera(s) 772, etc.), as described herein.
Cameras with a field of view that include portions of the interior environment within the cabin of the vehicle 700 (e.g., one or more OMS sensor(s) 701) may be used as part of an occupant monitoring system (OMS) such as, but not limited to, a driver monitoring system (DMS). For example, OMS sensors (e.g., the OMS sensor(s) 701) may be used (e.g., by the controller(s) 736) to track an occupant's and/or driver's gaze direction, head pose, and/or blinking. This gaze information may be used to determine a level of attentiveness of the occupant or driver (e.g., to detect drowsiness, fatigue, and/or distraction), and/or to take responsive action to prevent harm to the occupant or operator. In some embodiments, data from OMS sensors may be used to allow gaze-controlled operations triggered by driver and/or non-driver occupants such as, but not limited to, adjusting cabin temperature and/or airflow, opening and closing windows, controlling cabin lighting, controlling entertainment systems, adjusting mirrors, adjusting seat positions, and/or other operations. In some embodiments, an OMS may be used for applications such as determining when objects and/or occupants have been left behind in a vehicle cabin (e.g., by detecting occupant presence after the driver exits the vehicle).
Each of the components, features, and systems of the vehicle 700 in
Although the bus 702 is described herein as being a CAN bus, this is not intended to be limiting. For example, in addition to, or alternatively from, the CAN bus, FlexRay and/or Ethernet may be used. Additionally, although a single line is used to represent the bus 702, this is not intended to be limiting. For example, there may be any number of busses 702, which may include one or more CAN busses, one or more FlexRay busses, one or more Ethernet busses, and/or one or more other types of busses using a different protocol. In some examples, two or more busses 702 may be used to perform different functions, and/or may be used for redundancy. For example, a first bus 702 may be used for collision avoidance functionality and a second bus 702 may be used for actuation control. In any example, each bus 702 may communicate with any of the components of the vehicle 700, and two or more busses 702 may communicate with the same components. In some examples, each SoC 704, each controller 736, and/or each computer within the vehicle may have access to the same input data (e.g., inputs from sensors of the vehicle 700), and may be connected to a common bus, such the CAN bus.
The vehicle 700 may include one or more controller(s) 736, such as those described herein with respect to
The vehicle 700 may include a system(s) on a chip (SoC) 704. The SoC 704 may include CPU(s) 706, GPU(s) 708, processor(s) 710, cache(s) 712, accelerator(s) 714, data store(s) 716, and/or other components and features not illustrated. The SoC(s) 704 may be used to control the vehicle 700 in a variety of platforms and systems. For example, the SoC(s) 704 may be combined in a system (e.g., the system of the vehicle 700) with an HD map 722 which may obtain map refreshes and/or updates via a network interface 724 from one or more servers (e.g., server(s) 778 of
The CPU(s) 706 may include a CPU cluster or CPU complex (alternatively referred to herein as a “CCPLEX”). The CPU(s) 706 may include multiple cores and/or L2 caches. For example, in some embodiments, the CPU(s) 706 may include eight cores in a coherent multi-processor configuration. In some embodiments, the CPU(s) 706 may include four dual-core clusters where each cluster has a dedicated L2 cache (e.g., a 2 MB L2 cache). The CPU(s) 706 (e.g., the CCPLEX) may be configured to support simultaneous cluster operation allowing any combination of the clusters of the CPU(s) 706 to be active at any given time.
The CPU(s) 706 may implement power management capabilities that include one or more of the following features: individual hardware blocks may be clock-gated automatically when idle to save dynamic power; each core clock may be gated when the core is not actively executing instructions due to execution of WFI/WFE instructions; each core may be independently power-gated; each core cluster may be independently clock-gated when all cores are clock-gated or power-gated; and/or each core cluster may be independently power-gated when all cores are power-gated. The CPU(s) 706 may further implement an enhanced algorithm for managing power states, where allowed power states and expected wakeup times are specified, and the hardware/microcode determines the best power state to enter for the core, cluster, and CCPLEX. The processing cores may support simplified power state entry sequences in software with the work offloaded to microcode.
The GPU(s) 708 may include an integrated GPU (alternatively referred to herein as an “iGPU”). The GPU(s) 708 may be programmable and may be efficient for parallel workloads. The GPU(s) 708, in some examples, may use an enhanced tensor instruction set. The GPU(s) 708 may include one or more streaming microprocessors, where each streaming microprocessor may include an L1 cache (e.g., an L1 cache with at least 96 KB storage capacity), and two or more of the streaming microprocessors may share an L2 cache (e.g., an L2 cache with a 512 KB storage capacity). In some embodiments, the GPU(s) 708 may include at least eight streaming microprocessors. The GPU(s) 708 may use compute application programming interface(s) (API(s)). In addition, the GPU(s) 708 may use one or more parallel computing platforms and/or programming models (e.g., NVIDIA's CUDA).
The GPU(s) 708 may be power-optimized for best performance in automotive and embedded use cases. For example, the GPU(s) 708 may be fabricated on a Fin field-effect transistor (FinFET). However, this is not intended to be limiting and the GPU(s) 708 may be fabricated using other semiconductor manufacturing processes. Each streaming microprocessor may incorporate a number of mixed-precision processing cores partitioned into multiple blocks. For example, and without limitation, 64 PF32 cores and 32 PF64 cores may be partitioned into four processing blocks. In such an example, each processing block may be allocated 16 FP32 cores, 8 FP64 cores, 16 INT32 cores, two mixed-precision NVIDIA TENSOR COREs for deep learning matrix arithmetic, an L0 instruction cache, a warp scheduler, a dispatch unit, and/or a 64 KB register file. In addition, the streaming microprocessors may include independent parallel integer and floating-point data paths to provide for efficient execution of workloads with a mix of computation and addressing calculations. The streaming microprocessors may include independent thread scheduling capability to allow finer-grain synchronization and cooperation between parallel threads. The streaming microprocessors may include a combined L1 data cache and shared memory unit in order to improve performance while simplifying programming.
The GPU(s) 708 may include a high bandwidth memory (HBM) and/or a 16 GB HBM2 memory subsystem to provide, in some examples, about 900 GB/second peak memory bandwidth. In some examples, in addition to, or alternatively from, the HBM memory, a synchronous graphics random-access memory (SGRAM) may be used, such as a graphics double data rate type five synchronous random-access memory (GDDR5).
The GPU(s) 708 may include unified memory technology including access counters to allow for more accurate migration of memory pages to the processor that accesses them most frequently, thereby improving efficiency for memory ranges shared between processors. In some examples, address translation services (ATS) support may be used to allow the GPU(s) 708 to access the CPU(s) 706 page tables directly. In such examples, when the GPU(s) 708 memory management unit (MMU) experiences a miss, an address translation request may be transmitted to the CPU(s) 706. In response, the CPU(s) 706 may look in its page tables for the virtual-to-physical mapping for the address and transmits the translation back to the GPU(s) 708. As such, unified memory technology may allow a single unified virtual address space for memory of both the CPU(s) 706 and the GPU(s) 708, thereby simplifying the GPU(s) 708 programming and porting of applications to the GPU(s) 708.
In addition, the GPU(s) 708 may include an access counter that may keep track of the frequency of access of the GPU(s) 708 to memory of other processors. The access counter may help ensure that memory pages are moved to the physical memory of the processor that is accessing the pages most frequently.
The SoC(s) 704 may include any number of cache(s) 712, including those described herein. For example, the cache(s) 712 may include an L3 cache that is available to both the CPU(s) 706 and the GPU(s) 708 (e.g., that is connected both the CPU(s) 706 and the GPU(s) 708). The cache(s) 712 may include a write-back cache that may keep track of states of lines, such as by using a cache coherence protocol (e.g., MEI, MESI, MSI, etc.). The L3 cache may include 4 MB or more, depending on the embodiment, although smaller cache sizes may be used.
The SoC(s) 704 may include an arithmetic logic unit(s) (ALU(s)) which may be leveraged in performing processing with respect to any of the variety of tasks or operations of the vehicle 700—such as processing DNNs. In addition, the SoC(s) 704 may include a floating point unit(s) (FPU(s))—or other math coprocessor or numeric coprocessor types—for performing mathematical operations within the system. For example, the SoC(s) 704 may include one or more FPUs integrated as execution units within a CPU(s) 706 and/or GPU(s) 708.
The SoC(s) 704 may include one or more accelerators 714 (e.g., hardware accelerators, software accelerators, or a combination thereof). For example, the SoC(s) 704 may include a hardware acceleration cluster that may include optimized hardware accelerators and/or large on-chip memory. The large on-chip memory (e.g., 4 MB of SRAM), may allow the hardware acceleration cluster to accelerate neural networks and other calculations. The hardware acceleration cluster may be used to complement the GPU(s) 708 and to off-load some of the tasks of the GPU(s) 708 (e.g., to free up more cycles of the GPU(s) 708 for performing other tasks). As an example, the accelerator(s) 714 may be used for targeted workloads (e.g., perception, convolutional neural networks (CNNs), etc.) that are stable enough to be amenable to acceleration. The term “CNN,” as used herein, may include all types of CNNs, including region-based or regional convolutional neural networks (RCNNs) and Fast RCNNs (e.g., as used for object detection).
The accelerator(s) 714 (e.g., the hardware acceleration cluster) may include a deep learning accelerator(s) (DLA). The DLA(s) may include one or more Tensor processing units (TPUs) that may be configured to provide an additional ten trillion operations per second for deep learning applications and inferencing. The TPUs may be accelerators configured to, and optimized for, performing image processing functions (e.g., for CNNs, RCNNs, etc.). The DLA(s) may further be optimized for a specific set of neural network types and floating point operations, as well as inferencing. The design of the DLA(s) may provide more performance per millimeter than a general-purpose GPU, and vastly exceeds the performance of a CPU. The TPU(s) may perform several functions, including a single-instance convolution function, supporting, for example, INT8, INT16, and FP16 data types for both features and weights, as well as post-processor functions.
The DLA(s) may quickly and efficiently execute neural networks, especially CNNs, on processed or unprocessed data for any of a variety of functions, including, for example and without limitation: a CNN for object identification and detection using data from camera sensors; a CNN for distance estimation using data from camera sensors; a CNN for emergency vehicle detection and identification and detection using data from microphones; a CNN for facial recognition and vehicle owner identification using data from camera sensors; and/or a CNN for security and/or safety related events.
The DLA(s) may perform any function of the GPU(s) 708, and by using an inference accelerator, for example, a designer may target either the DLA(s) or the GPU(s) 708 for any function. For example, the designer may focus processing of CNNs and floating point operations on the DLA(s) and leave other functions to the GPU(s) 708 and/or other accelerator(s) 714.
The accelerator(s) 714 (e.g., the hardware acceleration cluster) may include a programmable vision accelerator(s) (PVA), which may alternatively be referred to herein as a computer vision accelerator. The PVA(s) may be designed and configured to accelerate computer vision algorithms for the advanced driver assistance systems (ADAS), autonomous driving, and/or augmented reality (AR) and/or virtual reality (VR) applications. The PVA(s) may provide a balance between performance and flexibility. For example, each PVA(s) may include, for example and without limitation, any number of reduced instruction set computer (RISC) cores, direct memory access (DMA), and/or any number of vector processors.
The RISC cores may interact with image sensors (e.g., the image sensors of any of the cameras described herein), image signal processor(s), and/or the like. Each of the RISC cores may include any amount of memory. The RISC cores may use any of a number of protocols, depending on the embodiment. In some examples, the RISC cores may execute a real-time operating system (RTOS). The RISC cores may be implemented using one or more integrated circuit devices, application specific integrated circuits (ASICs), and/or memory devices. For example, the RISC cores may include an instruction cache and/or a tightly coupled RAM.
The DMA may allow components of the PVA(s) to access the system memory independently of the CPU(s) 706. The DMA may support any number of features used to provide optimization to the PVA including, but not limited to, supporting multi-dimensional addressing and/or circular addressing. In some examples, the DMA may support up to six or more dimensions of addressing, which may include block width, block height, block depth, horizontal block stepping, vertical block stepping, and/or depth stepping.
The vector processors may be programmable processors that may be designed to efficiently and flexibly execute programming for computer vision algorithms and provide signal processing capabilities. In some examples, the PVA may include a PVA core and two vector processing subsystem partitions. The PVA core may include a processor subsystem, DMA engine(s) (e.g., two DMA engines), and/or other peripherals. The vector processing subsystem may operate as the primary processing engine of the PVA, and may include a vector processing unit (VPU), an instruction cache, and/or vector memory (e.g., VMEM). A VPU core may include a digital signal processor such as, for example, a single instruction, multiple data (SIMD), very long instruction word (VLIW) digital signal processor. The combination of the SIMD and VLIW may enhance throughput and speed.
Each of the vector processors may include an instruction cache and may be coupled to dedicated memory. As a result, in some examples, each of the vector processors may be configured to execute independently of the other vector processors. In other examples, the vector processors that are included in a particular PVA may be configured to employ data parallelism. For example, in some embodiments, the plurality of vector processors included in a single PVA may execute the same computer vision algorithm, but on different regions of an image. In other examples, the vector processors included in a particular PVA may simultaneously execute different computer vision algorithms, on the same image, or even execute different algorithms on sequential images or portions of an image. Among other things, any number of PVAs may be included in the hardware acceleration cluster and any number of vector processors may be included in each of the PVAs. In addition, the PVA(s) may include additional error correcting code (ECC) memory, to enhance overall system safety.
The accelerator(s) 714 (e.g., the hardware acceleration cluster) may include a computer vision network on-chip and SRAM, for providing a high-bandwidth, low latency SRAM for the accelerator(s) 714. In some examples, the on-chip memory may include at least 4 MB SRAM, consisting of, for example and without limitation, eight field-configurable memory blocks, that may be accessible by both the PVA and the DLA. Each pair of memory blocks may include an advanced peripheral bus (APB) interface, configuration circuitry, a controller, and a multiplexer. Any type of memory may be used. The PVA and DLA may access the memory via a backbone that provides the PVA and DLA with high-speed access to memory. The backbone may include a computer vision network on-chip that interconnects the PVA and the DLA to the memory (e.g., using the APB).
The computer vision network on-chip may include an interface that determines, before transmission of any control signal/address/data, that both the PVA and the DLA provide ready and valid signals. Such an interface may provide for separate phases and separate channels for transmitting control signals/addresses/data, as well as burst-type communications for continuous data transfer. This type of interface may comply with ISO 26262 or IEC 61508 standards, although other standards and protocols may be used.
In some examples, the SoC(s) 704 may include a real-time ray-tracing hardware accelerator, such as described in U.S. patent application Ser. No. 16/101,232, filed on Aug. 10, 2018. The real-time ray-tracing hardware accelerator may be used to quickly and efficiently determine the positions and extents of objects (e.g., within a world model), to generate real-time visualization simulations, for RADAR signal interpretation, for sound propagation synthesis and/or analysis, for simulation of SONAR systems, for general wave propagation simulation, for comparison to LiDAR data for purposes of localization and/or other functions, and/or for other uses. In some embodiments, one or more tree traversal units (TTUs) may be used for executing one or more ray-tracing related operations.
The accelerator(s) 714 (e.g., the hardware accelerator cluster) have a wide array of uses for autonomous driving. The PVA may be a programmable vision accelerator that may be used for key processing stages in ADAS and autonomous vehicles. The PVA's capabilities are a good match for algorithmic domains needing predictable processing, at low power and low latency. As such, the PVA performs well on semi-dense or dense regular computation, even on small data sets, which need predictable run-times with low latency and low power. Thus, in the context of platforms for autonomous vehicles, the PVAs are designed to run classic computer vision algorithms, as they are efficient at object detection and operating on integer math.
For example, according to one embodiment of the technology, the PVA is used to perform computer stereo vision. A semi-global matching-based algorithm may be used in some examples, although this is not intended to be limiting. Many applications for Level 3-5 autonomous driving require motion estimation/stereo matching on-the-fly (e.g., structure from motion, pedestrian recognition, lane detection, etc.). The PVA may perform computer stereo vision function on inputs from two monocular cameras.
In some examples, the PVA may be used to perform dense optical flow. According to process raw RADAR data (e.g., using a 4D Fast Fourier Transform) to provide Processed RADAR. In other examples, the PVA is used for time of flight depth processing, by processing raw time of flight data to provide processed time of flight data, for example.
The DLA may be used to run any type of network to enhance control and driving safety, including for example, a neural network that outputs a measure of confidence for each object detection. Such a confidence value may be interpreted as a probability, or as providing a relative “weight” of each detection compared to other detections. This confidence value enables the system to make further decisions regarding which detections should be considered as true positive detections rather than false positive detections. For example, the system may set a threshold value for the confidence and consider only the detections exceeding the threshold value as true positive detections. In an automatic emergency braking (AEB) system, false positive detections would cause the vehicle to automatically perform emergency braking, which is obviously undesirable. Therefore, only the most confident detections should be considered as triggers for AEB. The DLA may run a neural network for regressing the confidence value. The neural network may take as its input at least some subset of parameters, such as bounding box dimensions, ground plane estimate obtained (e.g. from another subsystem), inertial measurement unit (IMU) sensor 766 output that correlates with the vehicle 700 orientation, distance, 3D location estimates of the object obtained from the neural network and/or other sensors (e.g., LiDAR sensor(s) 764 or RADAR sensor(s) 760), among others.
The SoC(s) 704 may include data store(s) 716 (e.g., memory). The data store(s) 716 may be on-chip memory of the SoC(s) 704, which may store neural networks to be executed on the GPU and/or the DLA. In some examples, the data store(s) 716 may be large enough in capacity to store multiple instances of neural networks for redundancy and safety. The data store(s) 716 may comprise L2 or L3 cache(s) 712. Reference to the data store(s) 716 may include reference to the memory associated with the PVA, DLA, and/or other accelerator(s) 714, as described herein.
The SoC(s) 704 may include one or more processor(s) 710 (e.g., embedded processors). The processor(s) 710 may include a boot and power management processor that may be a dedicated processor and subsystem to handle boot power and management functions and related security enforcement. The boot and power management processor may be a part of the SoC(s) 704 boot sequence and may provide runtime power management services. The boot power and management processor may provide clock and voltage programming, assistance in system low power state transitions, management of SoC(s) 704 thermals and temperature sensors, and/or management of the SoC(s) 704 power states. Each temperature sensor may be implemented as a ring-oscillator whose output frequency is proportional to temperature, and the SoC(s) 704 may use the ring-oscillators to detect temperatures of the CPU(s) 706, GPU(s) 708, and/or accelerator(s) 714. If temperatures are determined to exceed a threshold, the boot and power management processor may enter a temperature fault routine and put the SoC(s) 704 into a lower power state and/or put the vehicle 700 into a chauffeur to safe stop mode (e.g., bring the vehicle 700 to a safe stop).
The processor(s) 710 may further include a set of embedded processors that may serve as an audio processing engine. The audio processing engine may be an audio subsystem that enables full hardware support for multi-channel audio over multiple interfaces, and a broad and flexible range of audio I/O interfaces. In some examples, the audio processing engine is a dedicated processor core with a digital signal processor with dedicated RAM.
The processor(s) 710 may further include an always on processor engine that may provide necessary hardware features to support low power sensor management and wake use cases. The always on processor engine may include a processor core, a tightly coupled RAM, supporting peripherals (e.g., timers and interrupt controllers), various I/O controller peripherals, and routing logic.
The processor(s) 710 may further include a safety cluster engine that includes a dedicated processor subsystem to handle safety management for automotive applications. The safety cluster engine may include two or more processor cores, a tightly coupled RAM, support peripherals (e.g., timers, an interrupt controller, etc.), and/or routing logic. In a safety mode, the two or more cores may operate in a lockstep mode and function as a single core with comparison logic to detect any differences between their operations.
The processor(s) 710 may further include a real-time camera engine that may include a dedicated processor subsystem for handling real-time camera management.
The processor(s) 710 may further include a high-dynamic range signal processor that may include an image signal processor that is a hardware engine that is part of the camera processing pipeline.
The processor(s) 710 may include a video image compositor that may be a processing block (e.g., implemented on a microprocessor) that implements video post-processing functions needed by a video playback application to produce the final image for the player window. The video image compositor may perform lens distortion correction on wide-view camera(s) 770, surround camera(s) 774, and/or on in-cabin monitoring camera sensors. In-cabin monitoring camera sensor is preferably monitored by a neural network running on another instance of the Advanced SoC, configured to identify in cabin events and respond accordingly. An in-cabin system may perform lip reading to activate cellular service and place a phone call, dictate emails, change the vehicle's destination, activate or change the vehicle's infotainment system and settings, or provide voice-activated web surfing. Certain functions are available to the driver only when the vehicle is operating in an autonomous mode, and are disabled otherwise.
The video image compositor may include enhanced temporal noise reduction for both spatial and temporal noise reduction. For example, where motion occurs in a video, the noise reduction weights spatial information appropriately, decreasing the weight of information provided by adjacent frames. Where an image or portion of an image does not include motion, the temporal noise reduction performed by the video image compositor may use information from the previous image to reduce noise in the current image.
The video image compositor may also be configured to perform stereo rectification on input stereo lens frames. The video image compositor may further be used for user interface composition when the operating system desktop is in use, and the GPU(s) 708 is not required to continuously render new surfaces. Even when the GPU(s) 708 is powered on and active doing 3D rendering, the video image compositor may be used to offload the GPU(s) 708 to improve performance and responsiveness.
The SoC(s) 704 may further include a mobile industry processor interface (MIPI) camera serial interface for receiving video and input from cameras, a high-speed interface, and/or a video input block that may be used for camera and related pixel input functions. The SoC(s) 704 may further include an input/output controller(s) that may be controlled by software and may be used for receiving I/O signals that are uncommitted to a specific role.
The SoC(s) 704 may further include a broad range of peripheral interfaces to allow communication with peripherals, audio codecs, power management, and/or other devices. The SoC(s) 704 may be used to process data from cameras (e.g., connected over Gigabit Multimedia Serial Link and Ethernet), sensors (e.g., LiDAR sensor(s) 764, RADAR sensor(s) 760, etc. that may be connected over Ethernet), data from bus 702 (e.g., speed of vehicle 700, steering wheel position, etc.), data from GNSS sensor(s) 758 (e.g., connected over Ethernet or CAN bus). The SoC(s) 704 may further include dedicated high-performance mass storage controllers that may include their own DMA engines, and that may be used to free the CPU(s) 706 from routine data management tasks.
The SoC(s) 704 may be an end-to-end platform with a flexible architecture that spans automation levels 3-5, thereby providing a comprehensive functional safety architecture that leverages and makes efficient use of computer vision and ADAS techniques for diversity and redundancy, provides a platform for a flexible, reliable driving software stack, along with deep learning tools. The SoC(s) 704 may be faster, more reliable, and even more energy-efficient and space-efficient than conventional systems. For example, the accelerator(s) 714, when combined with the CPU(s) 706, the GPU(s) 708, and the data store(s) 716, may provide for a fast, efficient platform for level 3-5 autonomous vehicles.
The technology thus provides capabilities and functionality that cannot be achieved by conventional systems. For example, computer vision algorithms may be executed on CPUs, which may be configured using high-level programming language, such as the C programming language, to execute a wide variety of processing algorithms across a wide variety of visual data. However, CPUs are oftentimes unable to meet the performance requirements of many computer vision applications, such as those related to execution time and power consumption, for example. In particular, many CPUs are unable to execute complex object detection algorithms in real-time, which is a requirement of in-vehicle ADAS applications, and a requirement for practical Level 3-5 autonomous vehicles.
In contrast to conventional systems, by providing a CPU complex, GPU complex, and a hardware acceleration cluster, the technology described herein allows for multiple neural networks to be performed simultaneously and/or sequentially, and for the results to be combined together to allow Level 3-5 autonomous driving functionality. For example, a CNN executing on the DLA or dGPU (e.g., the GPU(s) 720) may include a text and word recognition, allowing the supercomputer to read and understand traffic signs, including signs for which the neural network has not been specifically trained. The DLA may further include a neural network that is able to identify, interpret, and provides semantic understanding of the sign, and to pass that semantic understanding to the path planning modules running on the CPU Complex.
As another example, multiple neural networks may be run simultaneously, as is required for Level 3, 4, or 5 driving. For example, a warning sign consisting of “Caution: flashing lights indicate icy conditions,” along with an electric light, may be independently or collectively interpreted by several neural networks. The sign itself may be identified as a traffic sign by a first deployed neural network (e.g., a neural network that has been trained), the text “Flashing lights indicate icy conditions” may be interpreted by a second deployed neural network, which informs the vehicle's path planning software (preferably executing on the CPU Complex) that when flashing lights are detected, icy conditions exist. The flashing light may be identified by operating a third deployed neural network over multiple frames, informing the vehicle's path-planning software of the presence (or absence) of flashing lights. All three neural networks may run simultaneously, such as within the DLA and/or on the GPU(s) 708.
In some examples, a CNN for facial recognition and vehicle owner identification may use data from camera sensors to identify the presence of an authorized driver and/or owner of the vehicle 700. The always on sensor processing engine may be used to unlock the vehicle when the owner approaches the driver door and turn on the lights, and, in security mode, to disable the vehicle when the owner leaves the vehicle. In this way, the SoC(s) 704 provide for security against theft and/or carjacking.
In another example, a CNN for emergency vehicle detection and identification may use data from microphones 796 to detect and identify emergency vehicle sirens. In contrast to conventional systems, that use general classifiers to detect sirens and manually extract features, the SoC(s) 704 use the CNN for classifying environmental and urban sounds, as well as classifying visual data. In a preferred embodiment, the CNN running on the DLA is trained to identify the relative closing speed of the emergency vehicle (e.g., by using the Doppler Effect). The CNN may also be trained to identify emergency vehicles specific to the local area in which the vehicle is operating, as identified by GNSS sensor(s) 758. Thus, for example, when operating in Europe the CNN will seek to detect European sirens, and when in the United States the CNN will seek to identify only North American sirens. Once an emergency vehicle is detected, a control program may be used to execute an emergency vehicle safety routine, slowing the vehicle, pulling over to the side of the road, parking the vehicle, and/or idling the vehicle, with the assistance of ultrasonic sensors 762, until the emergency vehicle(s) passes.
The vehicle may include a CPU(s) 718 (e.g., discrete CPU(s), or dCPU(s)), that may be coupled to the SoC(s) 704 via a high-speed interconnect (e.g., PCIe). The CPU(s) 718 may include an X86 processor, for example. The CPU(s) 718 may be used to perform any of a variety of functions, including arbitrating potentially inconsistent results between ADAS sensors and the SoC(s) 704, and/or monitoring the status and health of the controller(s) 736 and/or infotainment SoC 730, for example.
The vehicle 700 may include a GPU(s) 720 (e.g., discrete GPU(s), or dGPU(s)), that may be coupled to the SoC(s) 704 via a high-speed interconnect (e.g., NVIDIA's NVLINK). The GPU(s) 720 may provide additional artificial intelligence functionality, such as by executing redundant and/or different neural networks, and may be used to train and/or update neural networks based on input (e.g., sensor data) from sensors of the vehicle 700.
The vehicle 700 may further include the network interface 724 which may include one or more wireless antennas 726 (e.g., one or more wireless antennas for different communication protocols, such as a cellular antenna, a Bluetooth antenna, etc.). The network interface 724 may be used to allow wireless connectivity over the Internet with the cloud (e.g., with the server(s) 778 and/or other network devices), with other vehicles, and/or with computing devices (e.g., client devices of passengers). To communicate with other vehicles, a direct link may be established between the two vehicles and/or an indirect link may be established (e.g., across networks and over the Internet). Direct links may be provided using a vehicle-to-vehicle communication link. The vehicle-to-vehicle communication link may provide the vehicle 700 information about vehicles in proximity to the vehicle 700 (e.g., vehicles in front of, on the side of, and/or behind the vehicle 700). This functionality may be part of a cooperative adaptive cruise control functionality of the vehicle 700.
The network interface 724 may include a SoC that provides modulation and demodulation functionality and enables the controller(s) 736 to communicate over wireless networks. The network interface 724 may include a radio frequency front-end for up-conversion from baseband to radio frequency, and down conversion from radio frequency to baseband. The frequency conversions may be performed through well-known processes, and/or may be performed using super-heterodyne processes. In some examples, the radio frequency front end functionality may be provided by a separate chip. The network interface may include wireless functionality for communicating over LTE, WCDMA, UMTS, GSM, CDMA2000, Bluetooth, Bluetooth LE, Wi-Fi, Z-Wave, ZigBee, LoRaWAN, and/or other wireless protocols.
The vehicle 700 may further include data store(s) 728 which may include off-chip (e.g., off the SoC(s) 704) storage. The data store(s) 728 may include one or more storage elements including RAM, SRAM, DRAM, VRAM, Flash, hard disks, and/or other components and/or devices that may store at least one bit of data.
The vehicle 700 may further include GNSS sensor(s) 758. The GNSS sensor(s) 758 (e.g., GPS, assisted GPS sensors, differential GPS (DGPS) sensors, etc.), to assist in mapping, perception, occupancy grid generation, and/or path planning functions. Any number of GNSS sensor(s) 758 may be used, including, for example and without limitation, a GPS using a USB connector with an Ethernet to Serial (RS-232) bridge.
The vehicle 700 may further include RADAR sensor(s) 760. The RADAR sensor(s) 760 may be used by the vehicle 700 for long-range vehicle detection, even in darkness and/or severe weather conditions. RADAR functional safety levels may be ASIL B. The RADAR sensor(s) 760 may use the CAN and/or the bus 702 (e.g., to transmit data generated using the RADAR sensor(s) 760) for control and to access object tracking data, with access to Ethernet to access raw data in some examples. A wide variety of RADAR sensor types may be used. For example, and without limitation, the RADAR sensor(s) 760 may be suitable for front, rear, and side RADAR use. In some example, Pulse Doppler RADAR sensor(s) are used.
The RADAR sensor(s) 760 may include different configurations, such as long range with narrow field of view, short range with wide field of view, short range side coverage, etc. In some examples, long-range RADAR may be used for adaptive cruise control functionality. The long-range RADAR systems may provide a broad field of view realized by two or more independent scans, such as within a 250 m range. The RADAR sensor(s) 760 may help in distinguishing between static and moving objects, and may be used by ADAS systems for emergency brake assist and forward collision warning. Long-range RADAR sensors may include monostatic multimodal RADAR with multiple (e.g., six or more) fixed RADAR antennae and a high-speed CAN and FlexRay interface. In an example with six antennae, the central four antennae may create a focused beam pattern, designed to record the vehicle's 700 surroundings at higher speeds with minimal interference from traffic in adjacent lanes. The other two antennae may expand the field of view, making it possible to quickly detect vehicles entering or leaving the vehicle's 700 lane.
Mid-range RADAR systems may include, as an example, a range of up to 760 m (front) or 80 m (rear), and a field of view of up to 42 degrees (front) or 750 degrees (rear). Short-range RADAR systems may include, without limitation, RADAR sensors designed to be installed at both ends of the rear bumper. When installed at both ends of the rear bumper, such a RADAR sensor systems may create two beams that constantly monitor the blind spot in the rear and next to the vehicle.
Short-range RADAR systems may be used in an ADAS system for blind spot detection and/or lane change assist.
The vehicle 700 may further include ultrasonic sensor(s) 762. The ultrasonic sensor(s) 762, which may be positioned at the front, back, and/or the sides of the vehicle 700, may be used for park assist and/or to create and update an occupancy grid. A wide variety of ultrasonic sensor(s) 762 may be used, and different ultrasonic sensor(s) 762 may be used for different ranges of detection (e.g., 2.5 m, 4 m). The ultrasonic sensor(s) 762 may operate at functional safety levels of ASIL B.
The vehicle 700 may include LiDAR sensor(s) 764. The LiDAR sensor(s) 764 may be used for object and pedestrian detection, emergency braking, collision avoidance, and/or other functions. The LiDAR sensor(s) 764 may be functional safety level ASIL B. In some examples, the vehicle 700 may include multiple LiDAR sensors 764 (e.g., two, four, six, etc.) that may use Ethernet (e.g., to provide data to a Gigabit Ethernet switch).
In some examples, the LiDAR sensor(s) 764 may be capable of providing a list of objects and their distances for a 360-degree field of view. Commercially available LiDAR sensor(s) 764 may have an advertised range of approximately 700 m, with an accuracy of 2 cm-3 cm, and with support for a 700 Mbps Ethernet connection, for example. In some examples, one or more non-protruding LiDAR sensors 764 may be used. In such examples, the LiDAR sensor(s) 764 may be implemented as a small device that may be embedded into the front, rear, sides, and/or corners of the vehicle 700. The LiDAR sensor(s) 764, in such examples, may provide up to a 120-degree horizontal and 35-degree vertical field-of-view, with a 200 m range even for low-reflectivity objects. Front-mounted LiDAR sensor(s) 764 may be configured for a horizontal field of view between 45 degrees and 135 degrees.
In some examples, LiDAR technologies, such as 3D flash LiDAR, may also be used. 3D Flash LiDAR uses a flash of a laser as a transmission source, to illuminate vehicle surroundings up to approximately 200 m. A flash LiDAR unit includes a receptor, which records the laser pulse transit time and the reflected light on each pixel, which in turn corresponds to the range from the vehicle to the objects. Flash LiDAR may allow for highly accurate and distortion-free images of the surroundings to be generated with every laser flash. In some examples, four flash LiDAR sensors may be deployed, one at each side of the vehicle 700. Available 3D flash LiDAR systems include a solid-state 3D staring array LiDAR camera with no moving parts other than a fan (e.g., a non-scanning LiDAR device). The flash LiDAR device may use a 5 nanosecond class I (eye-safe) laser pulse per frame and may capture the reflected laser light in the form of 3D range point clouds and co-registered intensity data. By using flash LiDAR, and because flash LiDAR is a solid-state device with no moving parts, the LiDAR sensor(s) 764 may be less susceptible to motion blur, vibration, and/or shock.
The vehicle may further include IMU sensor(s) 766. The IMU sensor(s) 766 may be located at a center of the rear axle of the vehicle 700, in some examples. The IMU sensor(s) 766 may include, for example and without limitation, an accelerometer(s), a magnetometer(s), a gyroscope(s), a magnetic compass(es), and/or other sensor types. In some examples, such as in six-axis applications, the IMU sensor(s) 766 may include accelerometers and gyroscopes, while in nine-axis applications, the IMU sensor(s) 766 may include accelerometers, gyroscopes, and magnetometers.
In some embodiments, the IMU sensor(s) 766 may be implemented as a miniature, high performance GPS-Aided Inertial Navigation System (GPS/INS) that combines micro-electro-mechanical systems (MEMS) inertial sensors, a high-sensitivity GPS receiver, and advanced Kalman filtering algorithms to provide estimates of position, velocity, and attitude. As such, in some examples, the IMU sensor(s) 766 may allow the vehicle 700 to estimate heading without requiring input from a magnetic sensor by directly observing and correlating the changes in velocity from GPS to the IMU sensor(s) 766. In some examples, the IMU sensor(s) 766 and the GNSS sensor(s) 758 may be combined in a single integrated unit.
The vehicle may include microphone(s) 796 placed in and/or around the vehicle 700. The microphone(s) 796 may be used for emergency vehicle detection and identification, among other things.
The vehicle may further include any number of camera types, including stereo camera(s) 768, wide-view camera(s) 770, infrared camera(s) 772, surround camera(s) 774, long-range and/or mid-range camera(s) 798, and/or other camera types. The cameras may be used to capture image data around an entire periphery of the vehicle 700. The types of cameras used depends on the embodiments and requirements for the vehicle 700, and any combination of camera types may be used to provide the necessary coverage around the vehicle 700. In addition, the number of cameras may differ depending on the embodiment. For example, the vehicle may include six cameras, seven cameras, ten cameras, twelve cameras, and/or another number of cameras. The cameras may support, as an example and without limitation, Gigabit Multimedia Serial Link (GMSL) and/or Gigabit Ethernet. Each of the camera(s) is described with more detail herein with respect to
The vehicle 700 may further include vibration sensor(s) 742. The vibration sensor(s) 742 may measure vibrations of components of the vehicle, such as the axle(s). For example, changes in vibrations may indicate a change in road surfaces. In another example, when two or more vibration sensors 742 are used, the differences between the vibrations may be used to determine friction or slippage of the road surface (e.g., when the difference in vibration is between a power-driven axle and a freely rotating axle).
The vehicle 700 may include an ADAS system 738. The ADAS system 738 may include a SoC, in some examples. The ADAS system 738 may include autonomous/adaptive/automatic cruise control (ACC), cooperative adaptive cruise control (CACC), forward crash warning (FCW), automatic emergency braking (AEB), lane departure warnings (LDW), lane keep assist (LKA), blind spot warning (BSW), rear cross-traffic warning (RCTW), collision warning systems (CWS), lane centering (LC), and/or other features and functionality.
The ACC systems may use RADAR sensor(s) 760, LiDAR sensor(s) 764, and/or a camera(s). The ACC systems may include longitudinal ACC and/or lateral ACC. Longitudinal ACC monitors and controls the distance to the vehicle immediately ahead of the vehicle 700 and automatically adjust the vehicle speed to maintain a safe distance from vehicles ahead. Lateral ACC performs distance keeping, and advises the vehicle 700 to change lanes when necessary. Lateral ACC is related to other ADAS applications such as LCA and CWS.
CACC uses information from other vehicles that may be received via the network interface 724 and/or the wireless antenna(s) 726 from other vehicles via a wireless link, or indirectly, over a network connection (e.g., over the Internet). Direct links may be provided by a vehicle-to-vehicle (V2V) communication link, while indirect links may be infrastructure-to-vehicle (I2V) communication link. In general, the V2V communication concept provides information about the immediately preceding vehicles (e.g., vehicles immediately ahead of and in the same lane as the vehicle 700), while the I2V communication concept provides information about traffic further ahead. CACC systems may include either or both I2V and V2V information sources. Given the information of the vehicles ahead of the vehicle 700, CACC may be more reliable and it has potential to improve traffic flow smoothness and reduce congestion on the road.
FCW systems are designed to alert the driver to a hazard, so that the driver may take corrective action. FCW systems use a front-facing camera and/or RADAR sensor(s) 760, coupled to a dedicated processor, DSP, FPGA, and/or ASIC, that is electrically coupled to driver feedback, such as a display, speaker, and/or vibrating component. FCW systems may provide a warning, such as in the form of a sound, visual warning, vibration and/or a quick brake pulse.
AEB systems detect an impending forward collision with another vehicle or other object, and may automatically apply the brakes if the driver does not take corrective action within a specified time or distance parameter. AEB systems may use front-facing camera(s) and/or RADAR sensor(s) 760, coupled to a dedicated processor, DSP, FPGA, and/or ASIC. When the AEB system detects a hazard, it typically first alerts the driver to take corrective action to avoid the collision and, if the driver does not take corrective action, the AEB system may automatically apply the brakes in an effort to prevent, or at least mitigate, the impact of the predicted collision. AEB systems, may include techniques such as dynamic brake support and/or crash imminent braking.
LDW systems provide visual, audible, and/or tactile warnings, such as steering wheel or seat vibrations, to alert the driver when the vehicle 700 crosses lane markings. A LDW system does not activate when the driver indicates an intentional lane departure, by activating a turn signal. LDW systems may use front-side facing cameras, coupled to a dedicated processor, DSP, FPGA, and/or ASIC, that is electrically coupled to driver feedback, such as a display, speaker, and/or vibrating component.
LKA systems are a variation of LDW systems. LKA systems provide steering input or braking to correct the vehicle 700 if the vehicle 700 starts to exit the lane.
BSW systems detects and warn the driver of vehicles in an automobile's blind spot. BSW systems may provide a visual, audible, and/or tactile alert to indicate that merging or changing lanes is unsafe. The system may provide an additional warning when the driver uses a turn signal. BSW systems may use rear-side facing camera(s) and/or RADAR sensor(s) 760, coupled to a dedicated processor, DSP, FPGA, and/or ASIC, that is electrically coupled to driver feedback, such as a display, speaker, and/or vibrating component.
RCTW systems may provide visual, audible, and/or tactile notification when an object is detected outside the rear-camera range when the vehicle 700 is backing up. Some RCTW systems include AEB to ensure that the vehicle brakes are applied to avoid a crash. RCTW systems may use one or more rear-facing RADAR sensor(s) 760, coupled to a dedicated processor, DSP, FPGA, and/or ASIC, that is electrically coupled to driver feedback, such as a display, speaker, and/or vibrating component.
Conventional ADAS systems may be prone to false positive results which may be annoying and distracting to a driver, but typically are not catastrophic, because the ADAS systems alert the driver and allow the driver to decide whether a safety condition truly exists and act accordingly. However, in an autonomous vehicle 700, the vehicle 700 itself must, in the case of conflicting results, decide whether to heed the result from a primary computer or a secondary computer (e.g., a first controller 736 or a second controller 736). For example, in some embodiments, the ADAS system 738 may be a backup and/or secondary computer for providing perception information to a backup computer rationality module. The backup computer rationality monitor may run a redundant diverse software on hardware components to detect faults in perception and dynamic driving tasks. Outputs from the ADAS system 738 may be provided to a supervisory MCU. If outputs from the primary computer and the secondary computer conflict, the supervisory MCU must determine how to reconcile the conflict to ensure safe operation.
In some examples, the primary computer may be configured to provide the supervisory MCU with a confidence score, indicating the primary computer's confidence in the chosen result. If the confidence score exceeds a threshold, the supervisory MCU may follow the primary computer's direction, regardless of whether the secondary computer provides a conflicting or inconsistent result. Where the confidence score does not meet the threshold, and where the primary and secondary computer indicate different results (e.g., the conflict), the supervisory MCU may arbitrate between the computers to determine the appropriate outcome.
The supervisory MCU may be configured to run a neural network(s) that is trained and configured to determine, based on outputs from the primary computer and the secondary computer, conditions under which the secondary computer provides false alarms. Thus, the neural network(s) in the supervisory MCU may learn when the secondary computer's output may be trusted, and when it cannot. For example, when the secondary computer is a RADAR-based FCW system, a neural network(s) in the supervisory MCU may learn when the FCW system is identifying metallic objects that are not, in fact, hazards, such as a drainage grate or manhole cover that triggers an alarm. Similarly, when the secondary computer is a camera-based LDW system, a neural network in the supervisory MCU may learn to override the LDW when bicyclists or pedestrians are present and a lane departure is, in fact, the safest maneuver. In embodiments that include a neural network(s) running on the supervisory MCU, the supervisory MCU may include at least one of a DLA or GPU suitable for running the neural network(s) with associated memory. In preferred embodiments, the supervisory MCU may comprise and/or be included as a component of the SoC(s) 704.
In other examples, ADAS system 738 may include a secondary computer that performs ADAS functionality using traditional rules of computer vision. As such, the secondary computer may use classic computer vision rules (if-then), and the presence of a neural network(s) in the supervisory MCU may improve reliability, safety and performance. For example, the diverse implementation and intentional non-identity makes the overall system more fault-tolerant, especially to faults caused by software (or software-hardware interface) functionality. For example, if there is a software bug or error in the software running on the primary computer, and the non-identical software code running on the secondary computer provides the same overall result, the supervisory MCU may have greater confidence that the overall result is correct, and the bug in software or hardware on primary computer is not causing material error.
In some examples, the output of the ADAS system 738 may be fed into the primary computer's perception block and/or the primary computer's dynamic driving task block. For example, if the ADAS system 738 indicates a forward crash warning due to an object immediately ahead, the perception block may use this information when identifying objects. In other examples, the secondary computer may have its own neural network which is trained and thus reduces the risk of false positives, as described herein.
The vehicle 700 may further include the infotainment SoC 730 (e.g., an in-vehicle infotainment system (IVI)). Although illustrated and described as a SoC, the infotainment system may not be a SoC, and may include two or more discrete components. The infotainment SoC 730 may include a combination of hardware and software that may be used to provide audio (e.g., music, a personal digital assistant, navigational instructions, news, radio, etc.), video (e.g., TV, movies, streaming, etc.), phone (e.g., hands-free calling), network connectivity (e.g., LTE, Wi-Fi, etc.), and/or information services (e.g., navigation systems, rear-parking assistance, a radio data system, vehicle related information such as fuel level, total distance covered, brake fuel level, oil level, door open/close, air filter information, etc.) to the vehicle 700. For example, the infotainment SoC 730 may radios, disk players, navigation systems, video players, USB and Bluetooth connectivity, carputers, in-car entertainment, Wi-Fi, steering wheel audio controls, hands free voice control, a heads-up display (HUD), an HMI display 734, a telematics device, a control panel (e.g., for controlling and/or interacting with various components, features, and/or systems), and/or other components. The infotainment SoC 730 may further be used to provide information (e.g., visual and/or audible) to a user(s) of the vehicle, such as information from the ADAS system 738, autonomous driving information such as planned vehicle maneuvers, trajectories, surrounding environment information (e.g., intersection information, vehicle information, road information, etc.), and/or other information.
The infotainment SoC 730 may include GPU functionality. The infotainment SoC 730 may communicate over the bus 702 (e.g., CAN bus, Ethernet, etc.) with other devices, systems, and/or components of the vehicle 700. In some examples, the infotainment SoC 730 may be coupled to a supervisory MCU such that the GPU of the infotainment system may perform some self-driving functions in the event that the primary controller(s) 736 (e.g., the primary and/or backup computers of the vehicle 700) fail. In such an example, the infotainment SoC 730 may put the vehicle 700 into a chauffeur to safe stop mode, as described herein.
The vehicle 700 may further include an instrument cluster 732 (e.g., a digital dash, an electronic instrument cluster, a digital instrument panel, etc.). The instrument cluster 732 may include a controller and/or supercomputer (e.g., a discrete controller or supercomputer). The instrument cluster 732 may include a set of instrumentation such as a speedometer, fuel level, oil pressure, tachometer, odometer, turn indicators, gearshift position indicator, seat belt warning light(s), parking-brake warning light(s), engine-malfunction light(s), airbag (SRS) system information, lighting controls, safety system controls, navigation information, etc. In some examples, information may be displayed and/or shared among the infotainment SoC 730 and the instrument cluster 732. As such, the instrument cluster 732 may be included as part of the infotainment SoC 730, or vice versa.
The server(s) 778 may receive, over the network(s) 790 and from the vehicles, image data representative of images showing unexpected or changed road conditions, such as recently commenced road-work. The server(s) 778 may transmit, over the network(s) 790 and to the vehicles, neural networks 792, updated neural networks 792, and/or map information 794, including information regarding traffic and road conditions. The updates to the map information 794 may include updates for the HD map 722, such as information regarding construction sites, potholes, detours, flooding, and/or other obstructions. In some examples, the neural networks 792, the updated neural networks 792, and/or the map information 794 may have resulted from new training and/or experiences represented in data received from any number of vehicles in the environment, and/or based on training performed at a datacenter (e.g., using the server(s) 778 and/or other servers).
The server(s) 778 may be used to train machine learning models (e.g., neural networks) based on training data. The training data may be generated using the vehicles, and/or may be generated in a simulation (e.g., using a game engine). In some examples, the training data is tagged (e.g., where the neural network benefits from supervised learning) and/or undergoes other pre-processing, while in other examples the training data is not tagged and/or pre-processed (e.g., where the neural network does not require supervised learning). Training may be executed according to any one or more classes of machine learning techniques, including, without limitation, classes such as: supervised training, semi-supervised training, unsupervised training, self-learning, reinforcement learning, federated learning, transfer learning, feature learning (including principal component and cluster analyses), multi-linear subspace learning, manifold learning, representation learning (including spare dictionary learning), rule-based machine learning, anomaly detection, and any variants or combinations therefor. Once the machine learning models are trained, the machine learning models may be used by the vehicles (e.g., transmitted to the vehicles over the network(s) 790, and/or the machine learning models may be used by the server(s) 778 to remotely monitor the vehicles.
In some examples, the server(s) 778 may receive data from the vehicles and apply the data to up-to-date real-time neural networks for real-time intelligent inferencing. The server(s) 778 may include deep-learning supercomputers and/or dedicated AI computers powered by GPU(s) 784, such as a DGX and DGX Station machines developed by NVIDIA. However, in some examples, the server(s) 778 may include deep learning infrastructure that use only CPU-powered datacenters.
The deep-learning infrastructure of the server(s) 778 may be capable of fast, real-time inferencing, and may use that capability to evaluate and verify the health of the processors, software, and/or associated hardware in the vehicle 700. For example, the deep-learning infrastructure may receive periodic updates from the vehicle 700, such as a sequence of images and/or objects that the vehicle 700 has located in that sequence of images (e.g., via computer vision and/or other machine learning object classification techniques). The deep-learning infrastructure may run its own neural network to identify the objects and compare them with the objects identified by the vehicle 700 and, if the results do not match and the infrastructure concludes that the AI in the vehicle 700 is malfunctioning, the server(s) 778 may transmit a signal to the vehicle 700 instructing a fail-safe computer of the vehicle 700 to assume control, notify the passengers, and complete a safe parking maneuver.
For inferencing, the server(s) 778 may include the GPU(s) 784 and one or more programmable inference accelerators (e.g., NVIDIA's TensorRT). The combination of GPU-powered servers and inference acceleration may make real-time responsiveness possible. In other examples, such as where performance is less critical, servers powered by CPUs, FPGAs, and other processors may be used for inferencing.
Although the various blocks of
The interconnect system 802 may represent one or more links or busses, such as an address bus, a data bus, a control bus, or a combination thereof. The interconnect system 802 may include one or more bus or link types, such as an industry standard architecture (ISA) bus, an extended industry standard architecture (EISA) bus, a video electronics standards association (VESA) bus, a peripheral component interconnect (PCI) bus, a peripheral component interconnect express (PCIe) bus, and/or another type of bus or link. In some embodiments, there are direct connections between components. As an example, the CPU 806 may be directly connected to the memory 804. Further, the CPU 806 may be directly connected to the GPU 808. Where there is direct, or point-to-point connection between components, the interconnect system 802 may include a PCIe link to carry out the connection. In these examples, a PCI bus need not be included in the computing device 800.
The memory 804 may include any of a variety of computer-readable media. The computer-readable media may be any available media that may be accessed by the computing device 800. The computer-readable media may include both volatile and nonvolatile media, and removable and non-removable media. By way of example, and not limitation, the computer-readable media may comprise computer-storage media and communication media.
The computer-storage media may include both volatile and nonvolatile media and/or removable and non-removable media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules, and/or other data types. For example, the memory 804 may store computer-readable instructions (e.g., that represent a program(s) and/or a program element(s), such as an operating system. Computer-storage media may include, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which may be used to store the desired information and which may be accessed by computing device 800. As used herein, computer storage media does not comprise signals per se.
The computer storage media may embody computer-readable instructions, data structures, program modules, and/or other data types in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media. The term “modulated data signal” may refer to a signal that has one or more of its characteristics set or changed in such a manner as to encode information in the signal. By way of example, and not limitation, the computer storage media may include wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, RF, infrared and other wireless media. Combinations of any of the above should also be included within the scope of computer-readable media.
The CPU(s) 806 may be configured to execute at least some of the computer-readable instructions to control one or more components of the computing device 800 to perform one or more of the methods and/or processes described herein. The CPU(s) 806 may each include one or more cores (e.g., one, two, four, eight, twenty-eight, seventy-two, etc.) that are capable of handling a multitude of software threads simultaneously. The CPU(s) 806 may include any type of processor, and may include different types of processors depending on the type of computing device 800 implemented (e.g., processors with fewer cores for mobile devices and processors with more cores for servers). For example, depending on the type of computing device 800, the processor may be an Advanced RISC Machines (ARM) processor implemented using Reduced Instruction Set Computing (RISC) or an x86 processor implemented using Complex Instruction Set Computing (CISC). The computing device 800 may include one or more CPUs 806 in addition to one or more microprocessors or supplementary co-processors, such as math co-processors.
In addition to or alternatively from the CPU(s) 806, the GPU(s) 808 may be configured to execute at least some of the computer-readable instructions to control one or more components of the computing device 800 to perform one or more of the methods and/or processes described herein. One or more of the GPU(s) 808 may be an integrated GPU (e.g., with one or more of the CPU(s) 806 and/or one or more of the GPU(s) 808 may be a discrete GPU. In embodiments, one or more of the GPU(s) 808 may be a coprocessor of one or more of the CPU(s) 806. The GPU(s) 808 may be used by the computing device 800 to render graphics (e.g., 3D graphics) or perform general purpose computations. For example, the GPU(s) 808 may be used for General-Purpose computing on GPUs (GPGPU). The GPU(s) 808 may include hundreds or thousands of cores that are capable of handling hundreds or thousands of software threads simultaneously. The GPU(s) 808 may generate pixel data for output images in response to rendering commands (e.g., rendering commands from the CPU(s) 806 received via a host interface). The GPU(s) 808 may include graphics memory, such as display memory, for storing pixel data or any other suitable data, such as GPGPU data. The display memory may be included as part of the memory 804. The GPU(s) 808 may include two or more GPUs operating in parallel (e.g., via a link). The link may directly connect the GPUs (e.g., using NVLINK) or may connect the GPUs through a switch (e.g., using NVSwitch). When combined together, each GPU 808 may generate pixel data or GPGPU data for different portions of an output or for different outputs (e.g., a first GPU for a first image and a second GPU for a second image). Each GPU may include its own memory, or may share memory with other GPUs.
In addition to or alternatively from the CPU(s) 806 and/or the GPU(s) 808, the logic unit(s) 820 may be configured to execute at least some of the computer-readable instructions to control one or more components of the computing device 800 to perform one or more of the methods and/or processes described herein. In embodiments, the CPU(s) 806, the GPU(s) 808, and/or the logic unit(s) 820 may discretely or jointly perform any combination of the methods, processes and/or portions thereof. One or more of the logic units 820 may be part of and/or integrated in one or more of the CPU(s) 806 and/or the GPU(s) 808 and/or one or more of the logic units 820 may be discrete components or otherwise external to the CPU(s) 806 and/or the GPU(s) 808. In embodiments, one or more of the logic units 820 may be a coprocessor of one or more of the CPU(s) 806 and/or one or more of the GPU(s) 808.
Examples of the logic unit(s) 820 include one or more processing cores and/or components thereof, such as Data Processing Units (DPUs), Tensor Cores (TCs), Tensor Processing Units(TPUs), Pixel Visual Cores (PVCs), Vision Processing Units (VPUs), Graphics Processing Clusters (GPCs), Texture Processing Clusters (TPCs), Streaming Multiprocessors (SMs), Tree Traversal Units (TTUs), Artificial Intelligence Accelerators (AIAs), Deep Learning Accelerators (DLAs), Arithmetic-Logic Units (ALUs), Application-Specific Integrated Circuits (ASICs), Floating Point Units (FPUs), input/output (I/O) elements, peripheral component interconnect (PCI) or peripheral component interconnect express (PCIe) elements, and/or the like.
The communication interface 810 may include one or more receivers, transmitters, and/or transceivers that allow the computing device 800 to communicate with other computing devices via an electronic communication network, included wired and/or wireless communications. The communication interface 810 may include components and functionality to allow communication over any of a number of different networks, such as wireless networks (e.g., Wi-Fi, Z-Wave, Bluetooth, Bluetooth LE, ZigBee, etc.), wired networks (e.g., communicating over Ethernet or InfiniBand), low-power wide-area networks (e.g., LoRaWAN, SigFox, etc.), and/or the Internet. In one or more embodiments, logic unit(s) 820 and/or communication interface 810 may include one or more data processing units (DPUs) to transmit data received over a network and/or through interconnect system 802 directly to (e.g., a memory of) one or more GPU(s) 808.
The I/O ports 812 may allow the computing device 800 to be logically coupled to other devices including the I/O components 814, the presentation component(s) 818, and/or other components, some of which may be built in to (e.g., integrated in) the computing device 800. Illustrative I/O components 814 include a microphone, mouse, keyboard, joystick, game pad, game controller, satellite dish, scanner, printer, wireless device, etc. The I/O components 814 may provide a natural user interface (NUI) that processes air gestures, voice, or other physiological inputs generated by a user. In some instances, inputs may be transmitted to an appropriate network element for further processing. An NUI may implement any combination of speech recognition, stylus recognition, facial recognition, biometric recognition, gesture recognition both on screen and adjacent to the screen, air gestures, head and eye tracking, and touch recognition (as described in more detail below) associated with a display of the computing device 800. The computing device 800 may be include depth cameras, such as stereoscopic camera systems, infrared camera systems, RGB camera systems, touchscreen technology, and combinations of these, for gesture detection and recognition. Additionally, the computing device 800 may include accelerometers or gyroscopes (e.g., as part of an inertia measurement unit (IMU)) that allow detection of motion. In some examples, the output of the accelerometers or gyroscopes may be used by the computing device 800 to render immersive augmented reality or virtual reality.
The power supply 816 may include a hard-wired power supply, a battery power supply, or a combination thereof. The power supply 816 may provide power to the computing device 800 to allow the components of the computing device 800 to operate.
The presentation component(s) 818 may include a display (e.g., a monitor, a touch screen, a television screen, a heads-up-display (HUD), other display types, or a combination thereof), speakers, and/or other presentation components. The presentation component(s) 818 may receive data from other components (e.g., the GPU(s) 808, the CPU(s) 806, DPUs, etc.), and output the data (e.g., as an image, video, sound, etc.).
As shown in
In at least one embodiment, grouped computing resources 914 may include separate groupings of node C.R.s 916 housed within one or more racks (not shown), or many racks housed in data centers at various geographical locations (also not shown). Separate groupings of node C.R.s 916 within grouped computing resources 914 may include grouped compute, network, memory or storage resources that may be configured or allocated to support one or more workloads. In at least one embodiment, several node C.R.s 916 including CPUs, GPUs, DPUs, and/or other processors may be grouped within one or more racks to provide compute resources to support one or more workloads. The one or more racks may also include any number of power modules, cooling modules, and/or network switches, in any combination.
The resource orchestrator 912 may configure or otherwise control one or more node C.R.s 916(1)-916(N) and/or grouped computing resources 914. In at least one embodiment, resource orchestrator 912 may include a software design infrastructure (SDI) management entity for the data center 900. The resource orchestrator 912 may include hardware, software, or some combination thereof.
In at least one embodiment, as shown in
In at least one embodiment, software 932 included in software layer 930 may include software used by at least portions of node C.R.s 916(1)-916(N), grouped computing resources 914, and/or distributed file system 938 of framework layer 920. One or more types of software may include, but are not limited to, Internet web page search software, e-mail virus scan software, database software, and streaming video content software.
In at least one embodiment, application(s) 942 included in application layer 940 may include one or more types of applications used by at least portions of node C.R.s 916(1)-916(N), grouped computing resources 914, and/or distributed file system 938 of framework layer 920. One or more types of applications may include, but are not limited to, any number of a genomics application, a cognitive compute, and a machine learning application, including training or inferencing software, machine learning framework software (e.g., PyTorch, TensorFlow, Caffe, etc.), and/or other machine learning applications used in conjunction with one or more embodiments.
In at least one embodiment, any of configuration manager 934, resource manager 936, and resource orchestrator 912 may implement any number and type of self-modifying actions based on any amount and type of data acquired in any technically feasible fashion. Self-modifying actions may relieve a data center operator of data center 900 from making possibly bad configuration decisions and possibly avoiding underutilized and/or poor performing portions of a data center.
The data center 900 may include tools, services, software or other resources to train one or more machine learning models or predict or infer information using one or more machine learning models according to one or more embodiments described herein. For example, a machine learning model(s) may be trained by calculating weight parameters according to a neural network architecture using software and/or computing resources described above with respect to the data center 900. In at least one embodiment, trained or deployed machine learning models corresponding to one or more neural networks may be used to infer or predict information using resources described above with respect to the data center 900 by using weight parameters calculated through one or more training techniques, such as but not limited to those described herein.
In at least one embodiment, the data center 900 may use CPUs, application-specific integrated circuits (ASICs), GPUs, FPGAs, and/or other hardware (or virtual compute resources corresponding thereto) to perform training and/or inferencing using above-described resources. Moreover, one or more software and/or hardware resources described above may be configured as a service to allow users to train or performing inferencing of information, such as image recognition, speech recognition, or other artificial intelligence services.
Network environments suitable for use in implementing embodiments of the disclosure may include one or more client devices, servers, network attached storage (NAS), other backend devices, and/or other device types. The client devices, servers, and/or other device types (e.g., each device) may be implemented on one or more instances of the computing device(s) 800 of
Components of a network environment may communicate with each other via a network(s), which may be wired, wireless, or both. The network may include multiple networks, or a network of networks. By way of example, the network may include one or more Wide Area Networks (WANs), one or more Local Area Networks (LANs), one or more public networks such as the Internet and/or a public switched telephone network (PSTN), and/or one or more private networks. Where the network includes a wireless telecommunications network, components such as a base station, a communications tower, or even access points (as well as other components) may provide wireless connectivity.
Compatible network environments may include one or more peer-to-peer network environments—in which case a server may not be included in a network environment—and one or more client-server network environments—in which case one or more servers may be included in a network environment. In peer-to-peer network environments, functionality described herein with respect to a server(s) may be implemented on any number of client devices.
In at least one embodiment, a network environment may include one or more cloud-based network environments, a distributed computing environment, a combination thereof, etc. A cloud-based network environment may include a framework layer, a job scheduler, a resource manager, and a distributed file system implemented on one or more of servers, which may include one or more core network servers and/or edge servers. A framework layer may include a framework to support software of a software layer and/or one or more application(s) of an application layer. The software or application(s) may respectively include web-based service software or applications. In embodiments, one or more of the client devices may use the web-based service software or applications (e.g., by accessing the service software and/or applications via one or more application programming interfaces (APIs)). The framework layer may be, but is not limited to, a type of free and open-source software web application framework such as that may use a distributed file system for large-scale data processing (e.g., “big data”).
A cloud-based network environment may provide cloud computing and/or cloud storage that carries out any combination of computing and/or data storage functions described herein (or one or more portions thereof). Any of these various functions may be distributed over multiple locations from central or core servers (e.g., of one or more data centers that may be distributed across a state, a region, a country, the globe, etc.). If a connection to a user (e.g., a client device) is relatively close to an edge server(s), a core server(s) may designate at least a portion of the functionality to the edge server(s). A cloud-based network environment may be private (e.g., limited to a single organization), may be public (e.g., available to many organizations), and/or a combination thereof (e.g., a hybrid cloud environment).
The client device(s) may include at least some of the components, features, and functionality of the example computing device(s) 800 described herein with respect to
The disclosure may be described in the general context of computer code or machine-useable instructions, including computer-executable instructions such as program modules, being executed by a computer or other machine, such as a personal data assistant or other handheld device. Generally, program modules including routines, programs, objects, components, data structures, etc., refer to code that perform particular tasks or implement particular abstract data types. The disclosure may be practiced in a variety of system configurations, including hand-held devices, consumer electronics, general-purpose computers, more specialty computing devices, etc. The disclosure may also be practiced in distributed computing environments where tasks are performed by remote-processing devices that are linked through a communications network.
As used herein, a recitation of “and/or” with respect to two or more elements should be interpreted to mean only one element, or a combination of elements. For example, “element A, element B, and/or element C” may include only element A, only element B, only element C, element A and element B, element A and element C, element B and element C, or elements A, B, and C. In addition, “at least one of element A or element B” may include at least one of element A, at least one of element B, or at least one of element A and at least one of element B. Further, “at least one of element A and element B” may include at least one of element A, at least one of element B, or at least one of element A and at least one of element B.
The subject matter of the present disclosure is described with specificity herein to meet statutory requirements. However, the description itself is not intended to limit the scope of this disclosure. Rather, the inventors have contemplated that the claimed subject matter might also be embodied in other ways, to include different steps or combinations of steps similar to the ones described in this document, in conjunction with other present or future technologies. Moreover, although the terms “step” and/or “block” may be used herein to connote different elements of methods employed, the terms should not be interpreted as implying any particular order among or between various steps herein disclosed unless and except when the order of individual steps is explicitly described.
The disclosure of this application also includes the following numbered clauses:
Clause 1. One or more processors comprising processing circuitry to apply two-dimensional (2D) sensor data generated using a source sensor rig of an ego-machine to at least a portion of a 3D perception network.
Clause 2. The one or more processors of clause 1, wherein the processing circuitry is further to apply simulated 2D sensor data to at least the portion of the 3D perception network, the simulated 2D sensor data being simulated based at least on a simulated source sensor rig and a simulated target sensor rig.
Clause 3. The one or more processors of clause 1 or 2, wherein the processing circuitry is further to update a three-dimensional (3D) perception network based at least on the 2D sensor data and the simulated 2D sensor data.
Clause 4. The one or more processors of clause 1, 2 or 3, wherein the processing circuitry is further to perform one or more operations with the ego-machine using the updated 3D perception network.
Clause 5. The one or processors of clause 1, 2, 3 or 4, wherein the processing circuitry is further to inject a representation of style extracted from the real 2D sensor data into one or more features extracted from the simulated 2D sensor data.
Clause 6. The one or processors of clause 1, 2, 3 or 4, wherein the processing circuitry is further to update the 3D perception network based at least on alternating between applying the real 2D sensor data to one or more first paths of at least the portion of the 3D perception network and applying the simulated 2D sensor data to one or more second paths of at least the portion of the 3D perception network.
Clause 7. The one or processors of clause 1, 2, 3 or 4, wherein the processing circuitry is further to update one or more shared weights shared by one or more first paths and one or more second paths of at least the portion of the 3D perception network based at least on alternating between applying the real 2D sensor data to the one or more first paths and applying the simulated 2D sensor data to the one or more second paths.
Clause 8. The one or processors of clause 1, 2, 3 or 4, wherein the processing circuitry is further to update the 3D perception network based at least on updating a viewpoint adjustment network comprising one or more layers of the 3D perception network shared across one or more first paths of the viewpoint adjustment network that process the real 2D sensor data and one or more second paths of the viewpoint adjustment network that process the simulated 2D sensor data.
Clause 9. The one or processors of clause 1, 2, 3 or 4, wherein the source sensor rig and the simulated source sensor rig represent a first sensor configuration of the ego-machine that differs from a second sensor configuration of the simulated target sensor rig.
Clause 10. The one or processors of clause 1, 2, 3 or 4, wherein the processing circuitry is further to generate, for at least one time slice of one or more time slices of a simulation, a simulated frame of 2D sensor data for at least one sensor of one or more sensors of the simulated source sensor rig and at least one corresponding sensor of the simulated target sensor rig and simulated ground truth data.
Clause 11. The one or processors of clause 1, 2, 3 or 4, wherein the processing circuitry is further to update a first viewpoint adjustment network comprising one or more layers of the 3D perception network based at least on applying the real 2D sensor data and the simulated 2D sensor data to the first viewpoint adjustment network, prior to updating a second viewpoint adjustment network comprising the one or more layers of the 3D perception network based at least on applying the simulated 2D sensor data to the second viewpoint adjustment network.
Clause 12. The one or processors of clause 1, 2, 3 or 4, wherein the 3D perception network is updated to perform at least one of a 2D-to-3D transformation or 2D-3D fusion.
Clause 13. The one or more processors of clause 1, 2, 3 or 4, wherein at least one of the 3D perception network or the one or more processors is comprised in at least one of: a control system for an autonomous or semi-autonomous machine; a perception system for an autonomous or semi-autonomous machine; a system for performing simulation operations; a system for performing digital twin operations; a system for performing light transport simulation; a system for performing collaborative content creation for 3D assets; a system for performing deep learning operations; a system for performing remote operations; a system for performing real-time streaming; a system for generating or presenting one or more of augmented reality content, virtual reality content, or mixed reality content; a system implemented using an edge device; a system implemented using a robot; a system for performing conversational AI operations; a system implementing one or more language models; a system implementing one or more large language models (LLMs); a system implementing one or more vision language models (VLMs); a system for generating synthetic data; a system for generating synthetic data using AI; a system incorporating one or more virtual machines (VMs); a system implemented at least partially in a data center; or a system implemented at least partially using cloud computing resources.
Clause 14. A system comprising one or more processors to perform one or more operations with an ego-machine using a three-dimensional (3D) perception network that is updated based at least on alternating between applying real two-dimensional (2D) sensor data generated using a source sensor rig of an ego-machine and applying simulated 2D sensor data to at least a portion of the 3D perception network, the simulated 2D sensor data being simulated based at least on a simulated source sensor rig and a simulated target sensor rig.
Clause 15. The system of clause 14, wherein the one or more processors are further to inject, based at least on applying the simulated 2D sensor data to at least the portion of the 3D perception network, a representation of style extracted from the real 2D sensor data into features extracted from the simulated 2D sensor data.
Clause 16. The system of clause 14, wherein the one or more processors are further to update the 3D perception network based at least on alternating between applying the real 2D sensor data to one or more first paths of at least the portion of the 3D perception network and applying the simulated 2D sensor data to one or more second paths of at least the portion of the 3D perception network.
Clause 17. The system of clause 14, wherein the one or more processors are further to update shared weights shared by one or more first paths and one or more second paths of at least the portion of the 3D perception network based at least on alternating between applying the real 2D sensor data to the one or more first paths and applying the simulated 2D sensor data to the one or more second paths.
Clause 18. The system of clause 14, wherein the one or more processors are further to update the 3D perception network based at least on updating a viewpoint adjustment network comprising one or more layers of the 3D perception network shared across one or more first paths of the viewpoint adjustment network that process the real 2D sensor data and one or more second paths of the viewpoint adjustment network that process the simulated 2D sensor data.
Clause 19. The system of clause 14, wherein the source sensor rig and the simulated source sensor rig represent a first sensor configuration of the ego-machine that differs from a second sensor configuration of the simulated target sensor rig.
Clause 20. The system of clause 14, wherein the one or more processors are further to generate, for at least one time slice of one or more time slices of a simulation, a simulated frame of 2D sensor data for at least one sensor of one or more sensors of the simulated source sensor rig and at least one corresponding sensor of the simulated target sensor rig and simulated ground truth data.
Clause 21. The system of clause 14, wherein at least one of the system or the 3D perception network is comprised in at least one of: a control system for an autonomous or semi-autonomous machine; a perception system for an autonomous or semi-autonomous machine; a system for performing simulation operations; a system for performing digital twin operations; a system for performing light transport simulation; a system for performing collaborative content creation for 3D assets; a system for performing deep learning operations; a system for performing remote operations; a system for performing real-time streaming; a system for generating or presenting one or more of augmented reality content, virtual reality content, or mixed reality content; a system implemented using an edge device; a system implemented using a robot; a system for performing conversational AI operations; a system implementing one or more language models; a system implementing one or more large language models (LLMs); a system implementing one or more vision language models (VLMs); a system for generating synthetic data; a system for generating synthetic data using AI; a system incorporating one or more virtual machines (VMs); a system implemented at least partially in a data center; or a system implemented at least partially using cloud computing resources.
Clause 22. A method comprising updating a three-dimensional (3D) reconstruction network based at least on alternating between: applying real two-dimensional (2D) sensor data generated using a first sensor rig of an ego-machine to at least a portion of the 3D reconstruction network.
Clause 23. The method of clause 22, further comprising and applying simulated 2D sensor data, generated based at least on simulating the first sensor rig and a second sensor rig, to at least the portion of the 3D reconstruction network.
Clause 24. The method of clause 22 or 23, further comprising performing one or more operations with the ego-machine using the 3D reconstruction network.
Clause 25. The method of clause 22, 23 or 24, wherein the method is performed by, or 3D reconstruction network is comprised in, at least one of: a control system for an autonomous or semi-autonomous machine; a perception system for an autonomous or semi-autonomous machine; a system for performing simulation operations; a system for performing digital twin operations; a system for performing light transport simulation; a system for performing collaborative content creation for 3D assets; a system for performing deep learning operations; a system for performing remote operations; a system for performing real-time streaming; a system for generating or presenting one or more of augmented reality content, virtual reality content, or mixed reality content; a system implemented using an edge device; a system implemented using a robot; a system for performing conversational AI operations; a system implementing one or more language models; a system implementing one or more large language models (LLMs); a system implementing one or more vision language models (VLMs); a system for generating synthetic data; a system for generating synthetic data using AI; a system incorporating one or more virtual machines (VMs); a system implemented at least partially in a data center; or a system implemented at least partially using cloud computing resources.
Clause 26. One or more processors comprising processing circuitry to perform one or more operations with an ego-machine using a three-dimensional (3D) perception network that is updated based at least on applying simulated two-dimensional (2D) sensor data to at least a portion of the 3D perception network.
Clause 27. The one or processors of clause 26, the simulated 2D sensor data being generated based at least on simulating a source sensor rig of an ego-machine and a target sensor rig.
Clause 28. The one or processors of clause 26 or 27, wherein the processing circuitry is further to update the 3D perception network based at least on updating a viewpoint adjustment network comprising one or more layers of the 3D perception network replicated as at least a portion of multiple channels of the viewpoint adjustment network.
Clause 29. The one or processors of clause 26 or 27, wherein the processing circuitry is further to update the 3D perception network based at least on applying the simulated 2D sensor data simulating the source sensor rig to one or more first paths of at least the portion of the 3D perception network and applying the simulated 2D sensor data simulating the target sensor rig to one or more second paths of at least the portion of the 3D perception network.
Clause 30. The one or processors of clause 26 or 27, wherein the 3D perception network comprises a view transformation that generates transformed features, wherein the processing circuitry is further to update the 3D perception network based at least on a loss that minimizes differences between the transformed features of one or more first paths and of one or more second paths of at least the portion of the 3D perception network.
Clause 31. The one or processors of clause 26 or 27, wherein the processing circuitry is further to update the 3D perception network based at least on a loss that minimizes differences between view-transformed features of one or more first paths that process the simulated 2D sensor data corresponding to the source sensor rig and one or more second paths that process the simulated 2D sensor data corresponding to the target sensor rig.
Clause 32. The one or processors of clause 26 or 27, wherein the source sensor rig represents a first sensor configuration that differs from a second sensor configuration of the target sensor rig.
Clause 33. The one or processors of clause 26 or 27, wherein the processing circuitry is further to generate, for at least one time slice of one or more time slices of a simulation, a simulated frame of 2D sensor data for at least one sensor of one or more simulated sensors of the source sensor rig and at least one corresponding sensor of the target sensor rig, and simulated ground truth data.
Clause 34. The one or processors of clause 26 or 27, wherein the processing circuitry is further to update a first viewpoint adjustment network comprising one or more layers of the 3D perception network based at least on applying real 2D sensor data and the simulated 2D sensor data to the first viewpoint adjustment network, prior to updating a second viewpoint adjustment network comprising the one or more layers of the 3D perception network based at least on applying the simulated 2D sensor data to the second viewpoint adjustment network.
Clause 35. The one or processors of clause 26 or 27, wherein the 3D perception network is updated to perform at least one of a 2D-to-3D transformation or 2D-3D fusion.
Clause 36. The one or more processors of clause 26 or 27, wherein at least one of the 3D perception network or the one or more processors is comprised in at least one of: a control system for an autonomous or semi-autonomous machine; a perception system for an autonomous or semi-autonomous machine; a system for performing simulation operations; a system for performing digital twin operations; a system for performing light transport simulation; a system for performing collaborative content creation for 3D assets; a system for performing deep learning operations; a system for performing remote operations; a system for performing real-time streaming; a system for generating or presenting one or more of augmented reality content, virtual reality content, or mixed reality content; a system implemented using an edge device; a system implemented using a robot; a system for performing conversational AI operations; a system implementing one or more language models; a system implementing one or more large language models (LLMs); a system implementing one or more vision language models (VLMs); a system for generating synthetic data; a system for generating synthetic data using AI; a system for generating synthetic data using generative AI; a system incorporating one or more virtual machines (VMs); a system implemented at least partially in a data center; or a system implemented at least partially using cloud computing resources.
Clause 37. A system comprising one or more processors to perform one or more operations with an ego machine using a three-dimensional (3D) reconstruction network that is updated based at least on applying simulated two-dimensional (2D) sensor data to at least a portion of the 3D reconstruction network.
Clause 38. The system of clause 37, the simulated 2D sensor data being generated based at least on simulating a source sensor rig of the ego-machine and a target sensor rig.
Clause 39. The system of clause 37 or 38, wherein the one or more processors are further to update the 3D reconstruction network based at least on updating a viewpoint adjustment network comprising one or more layers of the 3D reconstruction network replicated as at least a portion of multiple channels of the viewpoint adjustment network.
Clause 40. The system of clause 37 or 38, wherein the one or more processors are further to update the 3D reconstruction network based at least on applying the simulated 2D sensor data simulating the source sensor rig to one or more first paths of at least the portion of the 3D reconstruction network and applying the simulated 2D sensor data simulating the target sensor rig to one or more second paths of at least the portion of the 3D reconstruction network.
Clause 41. The system of clause 37 or 38, wherein the 3D reconstruction network is updated to perform a view transformation that generates transformed features, wherein the one or more processors are further to update the 3D reconstruction network based at least on a loss that minimizes differences between the transformed features of one or more first paths and of one or more second paths of at least the portion of the 3D reconstruction network.
Clause 42. The system of clause 37 or 38, wherein the one or more processors are further to update the 3D reconstruction network based at least on a loss that minimizes differences between view-transformed features of one or more first paths that process the simulated 2D sensor data corresponding to the source sensor rig and one or more second paths that process the simulated 2D sensor data corresponding to the target sensor rig.
Clause 43. The system of clause 37 or 38, wherein the source sensor rig represents a first sensor configuration that differs from a second sensor configuration of the target sensor rig.
Clause 44. The system of clause 37 or 38, wherein the one or more processors are further to generate, for at least one time slice of one or more time slices of a simulation, a simulated frame of 2D sensor data for at least one sensor of one or more simulated sensors of the source sensor rig and at least one corresponding sensor of the target sensor rig, and simulated ground truth data.
Clause 45. The system of clause 37 or 38, wherein at least one of the system or the 3D reconstruction network is comprised in at least one of: a control system for an autonomous or semi-autonomous machine; a perception system for an autonomous or semi-autonomous machine; a system for performing simulation operations; a system for performing digital twin operations; a system for performing light transport simulation; a system for performing collaborative content creation for 3D assets; a system for performing deep learning operations; a system for performing remote operations; a system for performing real-time streaming; a system for generating or presenting one or more of augmented reality content, virtual reality content, or mixed reality content; a system implemented using an edge device; a system implemented using a robot; a system for performing conversational AI operations; a system implementing one or more language models; a system implementing one or more large language models (LLMs); a system implementing one or more vision language models (VLMs); a system for generating synthetic data; a system for generating synthetic data using AI; a system for generating synthetic data using generative AI; a system incorporating one or more virtual machines (VMs); a system implemented at least partially in a data center; or a system implemented at least partially using cloud computing resources.
Clause 46. A method comprising applying simulated two-dimensional (2D) sensor data to at least a portion of the 3D reconstruction network.
Clause 47. The method of clause 46, the simulated 2D sensor data being generated based at least on simulating a first sensor rig of a first ego-machine and a second sensor rig of a second ego-machine.
Clause 48. The method of clause 46 or 47, further comprising updating a three-dimensional (3D) reconstruction network based at least on the simulated 2D sensor data.
Clause 49. The method of clause 46, 47 or 48, further comprising performing one or more operations with the second ego-machine using the 3D reconstruction network.
Clause 50. The method of clause 46, 47, 48 or 49, wherein the method is performed by, or the 3D reconstruction network is comprised in, at least one of: a control system for an autonomous or semi-autonomous machine; a perception system for an autonomous or semi-autonomous machine; a system for performing simulation operations; a system for performing digital twin operations; a system for performing light transport simulation; a system for performing collaborative content creation for 3D assets; a system for performing deep learning operations; a system for performing remote operations; a system for performing real-time streaming; a system for generating or presenting one or more of augmented reality content, virtual reality content, or mixed reality content; a system implemented using an edge device; a system implemented using a robot; a system for performing conversational AI operations; a system implementing one or more language models; a system implementing one or more large language models (LLMs); a system implementing one or more vision language models (VLMs); a system for generating synthetic data; a system for generating synthetic data using AI; a system for generating synthetic data using generative AI; a system incorporating one or more virtual machines (VMs); a system implemented at least partially in a data center; or a system implemented at least partially using cloud computing resources.
This application claims the benefit of U.S. Provisional Application No. 63/307,699, filed on Nov. 17, 2023, the contents of which are incorporated by reference in their entirety.
| Number | Date | Country | |
|---|---|---|---|
| 63600567 | Nov 2023 | US |
