Patent Application
20210261154
The present disclosure relates to the technical field of vehicles and, more specifically, to a continuous obstacle detection method, device, and system, and a storage medium.
With the development of driving assistance technology and automatic driving technology, millimeter wave radars are increasingly used in vehicles. Millimeter wave radar has the advantage of being able to operate in all-weather, with long operating range and high speed measurement accuracy, etc., which makes up for the deficiencies of other sensors such as ultrasound sensors and cameras. In conventional technology, a vehicle can be provided with a millimeter wave radar, and the millimeter wave radar can be used to detect the environment around the vehicle.
However, there is currently no technology that can detect continuous obstacles around the vehicle using millimeter wave radar. Generally, the millimeter wave radar presents the obstacles it detects in the form of dots. This form of presentation cannot determine the continuous obstacles around the vehicle, such as road edges, guardrails, fences, continuous stone posts, etc. Therefore, it is necessary to provide a method to enable the millimeter wave radar to determine such boundary characteristics on the road.
One aspect of the present disclosure provides a continuous obstacle detection method applied to a vehicle having a radar including an antenna configured to receive an echo signal. The method includes obtaining the echo signal at a current instance, and generating detection data at the current instance based on the echo signal; determining a plurality of stationary target points detected by the radar at the current instance based on the detection data at the current instance and vehicle information of the vehicle at the current instance; and determining a continuous obstacle trajectory at the current instance based on the plurality of stationary target points detected by the radar at the current instance.
Another aspect of the present disclosure provides a continuous obstacle detection system. The system includes a radar disposed on a vehicle, the radar including an antenna, the antenna being configured to receive an echo signal; a processor; and a memory storing one or more sets of computer program instruction sets that. When executed by the processor, the one or more sets of computer program instructions causes the processor to obtain the echo signal at a current instance, and generate detection data at the current instance based on the echo signal; determine a plurality of stationary target points detected by the radar at the current instance based on the detection data at the current instance and vehicle information of the vehicle at the current instance; and determine a continuous obstacle trajectory at the current instance based on the plurality of stationary target points detected by the radar at the current instance.
In order to illustrate the technical solutions in accordance with the embodiments of the present disclosure more clearly, the accompanying drawings to be used for describing the embodiments are introduced briefly in the following. It is apparent that the accompanying drawings in the following description are only some embodiments of the present disclosure. Persons of ordinary skill in the art can obtain other accompanying drawings in accordance with the accompanying drawings without any creative efforts.
Technical solutions of the present disclosure will be described in detail with reference to the drawings. It will be appreciated that the described embodiments represent some, rather than all, of the embodiments of the present disclosure. Other embodiments conceived or derived by those having ordinary skills in the art based on the described embodiments without inventive efforts should fall within the scope of the present disclosure.
It should be noted that, when one component is referred to as “fixed to” another component, it may be directly on another component or it is also possible that there is a third component between them. When one component is considered to “connect” another component, it may be directly connected to the other component or it is possible that there is a third component between them.
Unless otherwise defined, all the technical and scientific terms used in the present disclosure have the same or similar meanings as generally understood by one of ordinary skill in the art. As described in the present disclosure, the terms used in the specification of the present disclosure are intended to describe example embodiments, instead of limiting the present disclosure. The term “and/or” as used herein includes any and all combinations of one or more related listed items.
Exemplary embodiments will be described with reference to the accompanying drawings. In the case where there is no conflict between the exemplary embodiments, the features of the following embodiments and examples may be combined with each other.
An embodiment of the present disclosure provides a continuous obstacle detection method, which can be applied to a vehicle. The vehicle can include a radar, the radar includes at least an antenna, and the antenna can be used to receive echo signals. In some embodiments, the radar may be a millimeter wave radar. As shown in
In this embodiment, the radar may be a frequency modulated continuous wave (FMCW) radar. The FMCW radar may include an antenna, a radio frequency front end, a modulation module, and a signal processing unit. In some embodiments, the radio frequency front end may be used to transmit the detection signal, which can be a linear frequency modulation continuous wave, that is, the frequency of the detection signal emitted by the FMCW radar can be linearly modulated. Specifically, the modulation module may be used to linearly modulate the frequency of the detection signal emitted by the FMCW radar. When the detection signal emitted by the FMCW radar is reflected by objects around the vehicle, the antenna of the FMCW radar will receive the echo signal reflected by the object. The signal processing unit of the FMCW radar can process the echo signal to obtain detection data. In some embodiments, the detection data may include one or more of the energy of a target point detected by the radar, or the distance, speed, and angle of the target point relative to the radar.
In some embodiments, the FMCW radar may also communicate with an onboard processor. When the antenna receives the echo signal, the signal processing unit of the FMCW radar can perform an analog-to-digital conversion on the echo signal, that is, perform a digital sampling on the echo signal. Further, the sampled echo signal can be sent to the onboard processor, and the sampled echo signal can be processed by the onboard processor to obtain the detection data.
In other embodiments, the signal processing unit of the FMCW radar may process the echo signal. After obtaining the detection data, the detection data may also be sent to the onboard processor.
Further, the signal processing unit of the FMCW radar or the onboard processor may also determine continuous obstacles in the lane where the vehicle is positioned based on the detection data. The continuous obstacles may be fences, guardrails, road shoulders, continuous stone posts, etc. on the lane.
In other embodiments, the execution body of the continuous obstacle detection method may not be limited. The execution body of the continuous obstacle detection method may be the signal processing unit of the radar, the processor of the vehicle, a device with data processing function other than the signal processing unit of the radar and the processor on the vehicle, such as a server 12 shown in
S201, obtaining the echo signal at the current time, and generating the detection data at the current time based on the echo signal.
The execution body of the method in this embodiment may be a signal processing unit of a radar, an onboard processor, or the server 12 shown in
More specifically, the antenna of the radar can receive echo signals in real time, and the signal processing unit of the radar can generate real-time detection data of the radar based on the echo signals received in real time by the antenna. For example, the signal processing unit can obtain the echo signal received by the antenna at the current time, and perform an analog-to-digital conversion on the echo signal, that is, perform a digital sampling of the echo signal. Then, a fast Fourier transformation (FFT) can be performed on the sampled echo signal. More specifically, the signal processing unit can perform a two-dimensional FFT on the sampled echo signal, that is, the velocity dimension FFT and the distance dimension FFT, to obtain the detection data at the current time. It can be understood that the echo signals received by the antenna at different times may be different. Therefore, the signal processing unit can generate detection data at different times based on the echo signals received by the antenna at different times. Since the target points detected by the radar at different times may be different, the detection data at different times may be different.
In some embodiments, the detection data may include one or more of the energy of the target point detected by the radar, or the distance, speed, and the angle of the target point relative to the radar. In some embodiments, the detection data may be a two-dimensional data composed of a distance dimension and a speed dimension.
In some embodiments, there may be more than one FMCW radar. For example, the FMCW radar may include a plurality of antennas. At the same time, each of the plurality of antennas may receive echo signals. The signal processing unit may perform the analog-to-digital conversion and the two-dimensional FFT on the echo signals received by each antenna to obtain the two-dimensional data composed of the distance dimension and the speed dimension corresponding to each antenna. Further, a multi-channel incoherent accumulation can be performed on the two-dimensional data composed of the distance dimension and the speed dimension corresponding to each antenna to obtain the detection data. In some embodiments, each antenna may correspond to one channel, and the detection data obtained after the multi-channel incoherent accumulation may still be two-dimensional data composed of the distance dimension and the speed dimension. In addition, the two-dimensional data obtained at different times may be different.
In this embodiment, the two-dimensional data may be an N×M matrix, that is, a matrix of N rows and M columns. As shown in
S202, determining the stationary target points detected by the radar at the current time based on the detection data at the current time and the vehicle information of the vehicle at the current time.
As shown in
In some embodiments, the vehicle information may include one or more of the speed, steering, or yaw rate of the vehicle.
The radar is generally connected to the electrical and electronic systems on the vehicle through a communication bus to obtain vehicle information, such as the vehicle's speed, steering, and yaw rate. For example, the radar can be connected to the vehicle through the controller area network (CAN) bus and obtain the vehicle information of the vehicle from the CAN bus. In addition, the radar can also process the echo signals received by the antenna to obtain vehicle information, such as the speed of the vehicle. In this embodiment, the radar may obtain the vehicle information through the CAN bus or through signal processing. For example, in the process of obtaining the vehicle speed, the obtained vehicle speed can also be filtered. The specific filtering process is as shown in
S401, determining whether the vehicle speed obtained from the CAN bus is timed out at the current time, if yes, proceed to the process at S402, otherwise, proceed to the process at S403.
For example, the radar may obtain the vehicle speed from the Can bus at the current time, and determine whether the acquisition of the vehicle speed from the CAN bus at the current time is timed out.
S402, performing signal processing on the echo signal at the current time to obtain the vehicle speed, and using the vehicle speed as a valid input.
If the radar obtains the vehicle speed from the CAN bus at the current time, it can process the echo signal at the current time to obtain the vehicle speed, and use the vehicle speed as a valid point.
S403, using the current vehicle speed obtained from the CAN bus as a valid input.
If the radar obtains the vehicle speed from the CAN bus at the current time is not timed out, the vehicle speed obtained from the CAN bus at the current time can be used as a valid input.
S404, calculating the error of the current vehicle speed.
S405, determining whether the error exceeds an error threshold, if yes, execute the process at S406, other, execute the process at S410.
S406, incrementing the counter by 1.
S407, determining whether the value of the counter is greater than or equal to N, if yes, execute the process at S408, otherwise, execute the process at S409.
S408, reporting an error and resetting the trusted vehicle speed.
S409, maintaining the trusted vehicle speed unchanged.
S410, clearing the counter.
S411, performing filtering processing on the current vehicle speed to update the trusted vehicle speed.
For example, based on the above processes at S401-S403, the corresponding relationship between the determined valid input of the vehicle speed at different times and the trusted vehicle speed at different times are shown in Table 1 or Table 2.
For example, the valid input of the vehicle speed at time T0 is 10.5, and the error of 105.5 is less than the error threshold, then 10.5 can be filtered to obtain the trusted vehicle speed at time T0, such as 10. The valid input of the vehicle speed at T1 is 10.3, and the error of 10.3 is less than the error threshold, then 10.3 can be filtered. More specifically, based on the trusted vehicle speed 10 at T0 and the value input 10.3 of the vehicle speed at T1, a trusted vehicle speed 10.2 at T1 can be calculated, and the counter at T1 can be cleared. Assume that N=3, the valid input of the vehicle speed at T2 is 13, and the error 13 exceeds the error threshold, then the counter can be incremented by 1. The counter value at T2 is 1, 1 is less than N, and the trusted speed of 10.2 remains unchanged. The valid input of the vehicle speed at T3 is 14, if the error of 14 exceeds the error threshold, the counter will be incremented by 1. The counter value at time T3 is 2, 2 is less than N, and the trusted speed of 10.2 can remain unchanged at this time. The valid input of the vehicle speed at T4 is 15, and the error of 15 exceeds the error threshold, then the counter will be incremented by 1. The counter value at T4 is 3, and 3 is equal to N. At this time, an error can be reported and the trusted vehicle speed can be reset. For example, the trusted vehicle speed can be reset to the valid input of the vehicle speed at T4, which is 15. The subsequent process ca be deduced by analogy and will not be repeated here.
In addition, if the value of the counter has not reached N and a vehicle speed with an error less than the error threshold is entered, the counter can be cleared, and the newly input vehicle speed can be filtered to obtain the new trusted vehicle speed. For example, the valid input of the vehicle speed at T4 is 10.4, and the error of 10.4 is less than the error threshold, the counter can be cleared and 10.4 can be filtered. Based on the valid input 10.2 of the trusted vehicle speed maintained at T3 and 10.4 of the vehicle speed at T4, the trusted vehicle speed at T4 can be calculated, such as 10.3. The details are shown in Table 2 below.
In some embodiments, the trusted vehicle speed at different times determined through the above processes can be used as the real vehicle speed at different times, and the real vehicle speed can be used to determine the stationary target points.
In some embodiments, determining the stationary target points detected by the radar at the current time based on the detection data at the current time and the vehicle information of the vehicle at the current time may include determining the stationary target points detected by the radar at the current time based on the speed of the vehicle at the current time, and the distance and speed of the target points detected by the radar at the current time relative to the radar.
As shown in
In some embodiments, determining the stationary target point detected by the radar at the current time based on the speed of the vehicle at the current time, and the distance and speed of the target points detected by the radar at the current time relative to the radar may include, if the distance of the target point relative to the radar detected by the radar at the current time is greater than a predetermined distance, comparing the speed of the target point relative to the radar with the speed of the vehicle at the current time; and, if the different between the speed of the target point relative to the radar and the speed of the vehicle at the current time is less than a first predetermined difference, determining the target point as the stationary target point detected by the radar at the current time.
Take the points in black in
In some embodiments, determining the stationary target point detected by the radar at the current time based on the speed of the vehicle at the current time, and the distance and speed of the target points detected by the radar at the current time relative to the radar may include, if the distance of the target point detected by the radar at the current time relative to the radar is less than or equal to the predetermined distance, determining the equivalent ground speed of the target point based on the angle of the target point relative to the radar; and comparing the equivalent ground speed of the target point with the speed of the vehicle at the current time; and, if the difference between the equivalent ground speed of the target point and the speed of the vehicle at the current is less than a second predetermined difference, determining the target point as the stationary target point detected by the radar at the current time.
As shown in
For example, after detecting each target point whose energy is greater than the predetermined energy threshold shown in
In some embodiments, based on
As shown in
In some other embodiments, if the number of stationary target points determined by the radar signal processing unit based on the echo signal received by the antenna at the current time is relatively limited, the signal processing unit may also perform a multi-frame accumulation on the stationary target points. For example, the signal processing unit determines seven stationary target points at time t1, and eight stationary target points at time t2 after time t1, based on the displacement of the vehicle from time t1 to time t2, the stationary target points at time t1 can be compensated, and the compensated stationary target points and the stationary target points determined at time t2 can be accumulated to increase the density of the stationary target points. For example, at time t1, the seven stationary target points are at 80 meters, 81 meters, 82 meters, 83 meters, 84 m meters 85 meters, and 86 meters in front of the vehicle. The vehicle moves 10 meters forward from time t1 to time t2, at time t2, the seven stationary target points are 70 meters, 71 meters, 72 meters, 73 meters, 74 m meters 75 meters, and 76 meters in front of the vehicle. Therefore, the seven stationary target points at time t1 after position compensation and the stationary target points determined at time t2 can be accumulated.
S203, determining a continuous obstacle trajectory at the current time based on the stationary target points detected by the radar at the current time.
Since continuous obstacles such as fences, guardrails, road shoulders, continuous stone posts, or green belts on the lane stationary objects relative to the ground, the stationary target points determined by the above processes can be used as the target points for fitting the continuous obstacles. More specifically, based on the stationary target points determined at the current time, the continuous obstacle trajectory at the current time can be determined.
In some embodiments, determining the continuous obstacle trajectory at the current time based on the stationary target points detected by the radar at the current time may include generating the continuous obstacle trajectory at the current time based on the stationary target points detected by the radar at the current time.
For example, in the initial stage of establishing the continuous obstacle trajectory, there may be no continuous obstacle trajectory in the previous time. At this time, the continuous obstacle trajectory at the current time may be generated based on the stationary target points detected by the radar at the current time.
In some embodiments, determining the continuous obstacle trajectory at the current time based on the stationary target points detected by the radar at the current time may include, based on the stationary target points detected by the radar at the current time, updating the continuous obstacle trajectory at a previous time to obtain the continuous obstacle trajectory at the current time.
For example, the continuous obstacle trajectory has been established at a previous time, and a new is determined based on the stationary target points detected by the radar at the current time, then the degree of match between the new trajectory and the continuous obstacle trajectory at the previous time can be calculated. If the degree of match between the new trajectory and the continuous obstacle trajectory at the previous time is greater than a predetermined degree of match, the new trajectory can be associated with the continuous obstacle trajectory at the previous time, thereby updating the continuous obstacle trajectory at the previous time and obtaining the continuous obstacle trajectory at the current time.
In some embodiments, after determining the continuous obstacle trajectory at the current time based on the stationary target points detected by the radar at the current time, the method may further include determining a boundary of the lane where the vehicle is positioned at the current time based on the continuous obstacle trajectory at the current time. The boundary of the lane where the vehicle is positioned may be further applied to the field of assisted driving or autonomous vehicle.
In this embodiment, the echo signal at the current time can be obtained, the detection data at the current time can be generated based on the echo signal, the stationary target points detected by the radar at the current time can be determined based on the detection data at the current time and the vehicle information of the vehicle at the current time, and continuous obstacle trajectory at the current time can be determined based on the stationary target points detected by the radar at the current time, thereby realizing the detection of continuous obstacles around the vehicle.
An embodiment of the present disclosure provides a continuous obstacle detection method.
S801, performing clustering processing on the stationary target points detected by the radar at the current time to obtain the cluster at the current time.
For example, based on
The present disclosure odes not limit the clustering algorithm used on the clustering processing. For example, the clustering algorithm may be a density-based clustering algorithm (density-based spatial clustering of applications with noise (DB SCAN), or ordering points to identify the clustering structure (OPTICS), or DENsity-based CLUstEring (DENCLUE)) or a random sample consensus algorithm.
S802, if the quality of the cluster at the current time is greater than a predetermined quality threshold, generating the continuous obstacle trajectory at the current time based on the cluster at the current time.
In some embodiments, if the quality of the cluster at the current time is greater than the predetermined quality threshold, then before generating the continuous obstacle trajectory at the current time based on the cluster at the current time, the method may further include determining the quality of the cluster at the current time based on one or more of the number of cluster points in the cluster at the current time, the length of the cluster, or the degree of match between the cluster and the vehicle information of the vehicle.
For example, the cluster 1 and the cluster 2 in
In some embodiments, the target points detected by the radar may be converted from the matrix composed of the distance dimension and the speed dimension to the vehicle's own coordinate system.
More specifically, generating the continuous obstacle trajectory at the current time based on the good cluster at the current time may include performing parameter fitting on the good cluster at the current time to obtain parameter information of the continuous obstacle trajectory at the current time. The parameter information can uniquely describe the continuous obstacle trajectory at the current time, and the schematic diagram of the continuous obstacle trajectory can be the dotted line 110 shown in
In some embodiments, the polynomial fitting can be used to fit the parameters of the good cluster at the current time, or the radius arc fitting method can be used to fit the parameters of the good cluster at the current time. In some embodiments, the polynomial fitting method can be divided into first-order fitting, second-order fitting, and third-order fitting. Take the second-order fitting as an example, the parameter information obtained after parameter fitting of the good cluster at the current time may include a zero-order coefficient, a first-order coefficient, a second-order coefficient, closest distance information, and farthest distance information, where the closest distance information and the farthest distance information may both refer to the distance relative to the radar or the vehicle. If the radius arc fitting method is used to perform the parameter fitting on the good cluster at the current time, the parameter information obtained after fitting may include a center position, a radius, a starting arc, and an ending arc.
In some embodiments, after generating the continuous obstacle trajectory at the current time based on the cluster at the current time, the continuous obstacle trajectory at the current time and/or the parameter information of the continuous obstacle trajectory may also be output. As shown in
In this embodiment, clustering processing can be performed on the stationary target points detected by the radar at the current time to obtain the cluster at the current time. If the quality of the cluster at the current time is greater than the predetermined quality threshold, the continuous obstacle trajectory at the current time can be generated based on the cluster at the current time, thereby realizing the method of establishing the continuous obstacle trajectory at the initial of detecting the continuous obstacle.
An embodiment of the present disclosure provides a continuous obstacle detection method.
S1201, performing clustering processing on the stationary target points detected by the radar at the current time to obtain the cluster at the current time.
As shown in
S1202, calculating the degree of match between the cluster points in the cluster at the current time and the continuous obstacle trajectory at the previous time.
In some embodiments, before calculating the degree of match between the cluster points in the cluster at the current time and the continuous obstacle trajectory at the previous time, the method may further include determining the quality of the cluster at the current time based on one or more of the number of cluster points in the cluster at the current time, the length of the cluster, and the degree of match between the cluster and the vehicle information of the vehicle. Correspondingly, calculating the degree of match between the cluster points in the cluster at the current time and the continuous obstacle trajectory at the previous time may include calculating the degree of match between the cluster points in the cluster at the current time and the continuous obstacle trajectory at the previous time if the quality of the cluster at the current time is greater than the predetermined quality threshold.
For example, after performing clustering processing on the stationary target points detected by the radar at the current time and obtain the cluster in the dotted line frame 130 as shown in
In some embodiments, calculating the degree of match between the cluster points in the cluster at the current time and the continuous obstacle trajectory at the previous time may include calculating the degree of match between the cluster points in the cluster at the current time and the continuous obstacle trajectory at the previous time based on the distance of the cluster points in the cluster relative to the radar at the current time and the distance of the previous trajectory points in the continuous obstacle trajectory at the previous time relative to the radar; and/or, calculating the degree of match between the cluster points in the cluster at the current time and the continuous obstacle trajectory at the previous time based on the speed of the cluster points in the cluster relative to the radar at the current time and the speed of the previous trajectory points in the continuous obstacle trajectory at the previous time relative to the radar.
For example, the average distance of each cluster point relative to the radar can be calculated based on the distance of each cluster point in the dotted line frame 130 relative to the radar. The average distance of each previous trajectory point relative to the radar can be calculated based on the continuous obstacle trajectory at the previous time, such as the distance of each previous trajectory point in the dotted line 110 relative to the radar. Further, the average distance of each cluster point relative to the radar in the dotted line frame 130 can be compared with the average distance of each previous trajectory point relative to the radar. If the difference between the two is less than a predetermined value, the degree of match between the cluster points in the dotted line frame 130 and the continuous obstacle trajectory at the previous time can be determined to be greater than a predetermined degree of match.
Alternatively, the continuous obstacle trajectory at the previous time, such as the plurality of previous trajectory points near the dotted line frame 130 in the dotted line 110, can be selected. The plurality of previous trajectory points are shown as the white points in
For example, the degree of match between the cluster points in the dotted line frame 130 and the continuous obstacle trajectory at the previous time can also be calculated based on the speed of each cluster point in the dotted line frame 130 relative to the radar and the continuous obstacle trajectory at the previous time, such as the speed of each history trajectory point in the dotted line 110 relative to the radar. More specifically, the calculation of the degree of match between the cluster points in the dotted line frame 130 and the continuous obstacle trajectory at the previous time may not be limited to comparing the average speed of each cluster point in the dotted line frame 130 relative to the radar and the continuous obstacle trajectory at the previous time, such as the average speed of each previous trajectory point in the dotted line 110 relative to the radar.
In some embodiments, calculating the degree of match between the cluster points in the cluster at the current time and the continuous obstacle trajectory at the previous time may include performing parameter fitting on the cluster at the current time to obtain the parameter information of the new trajectory corresponding to the cluster at the current time; and calculating the degree of match between the cluster points at the current time and the continuous obstacle trajectory at the previous time based on the parameter information of the new trajectory corresponding to the cluster at the current time and the parameter of the continuous obstacle trajectory at the previous time.
As shown in
For example, take the second-order fitting as an example, after the second-order fitting method is used to perform parameter fitting on the cluster in the dotted line frame 130, the parameter information corresponding to the curve 150 can be obtained. Further, based on the parameter information corresponding to the curve 150 and the continuous obstacle trajectory at the previous time, such as the parameter information of the dotted line 110, the degree of match between the cluster points in the dotted line frame 130 and the continuous obstacle trajectory at the previous time can be calculated. For example, the parameter information corresponding to the curve 150 may include a zero-order coefficient, a first-order coefficient, a second-order coefficient, closest distance information, and farthest distance information, and the parameter information of the continuous obstacle trajectory at the previous time may also include a zero-order coefficient, a first-order coefficient, a second-order coefficient, closest distance information, and farthest distance information. Sequentially calculate the difference between the zero-order coefficient corresponding to the curve 150 and the zero-order coefficient corresponding to the continuous obstacle trajectory at the previous time, the difference between the first-order coefficient corresponding to the curve 150 and the first-order coefficient corresponding to the continuous obstacle trajectory at the previous time, the difference between the second-order coefficient corresponding to the curve 150 and the second-order coefficient corresponding to the continuous obstacle trajectory at the previous time, the difference between the closest distance corresponding to the curve 150 and the closest distance corresponding to the continuous obstacle trajectory at the previous time, and the difference between the farther distance corresponding to the curve 150 and the farther distance corresponding to the continuous obstacle trajectory at the previous time. If the aforementioned differences are all within a predetermined range, the degree of match between the cluster points in the dotted line frame 130 and the continuous obstacle trajectory at the previous time may be determined to be greater than the predetermined degree of match.
S1203, associating the cluster points in the cluster at the current time with the continuous obstacle trajectory at the previous time to obtain the continuous obstacle trajectory at the current time if the degree of match between the cluster points in the cluster at the current time and the continuous obstacle trajectory at the previous time is greater than the predetermined degree of match.
For example, if the degree of match between the cluster points in the dotted line frame 130 and the continuous obstacle trajectory at the previous time is greater than the predetermined degree of match, then the cluster points in the dotted line frame 130 and the continuous obstacle trajectory at the previous time, such as the dotted line 110, can be associated to obtained the continuous obstacle trajectory at the current time.
In some embodiments, associating the cluster points in the cluster at the current time with the continuous obstacle trajectory at the previous time to obtain the continuous obstacle trajectory at the current time may include obtaining the new trajectory corresponding to the cluster at the current time based on the cluster points in the cluster at the current time; and, associating the new trajectory corresponding to the cluster at the current time with the continuous obstacle trajectory at the previous time to obtain the continuous obstacle trajectory at the current time, the continuous obstacle trajectory at the current time including the new trajectory and the continuous obstacle trajectory at the previous time.
For example, the parameter information of the curve 150 can be obtained by performing parameter fitting on the cluster points in the dotted line frame 130. The curve 150 may be the new trajectory obtained from the cluster points in the dotted line frame 130, and the new trajectory may be associated with the continuous obstacle trajectory at the previous time, such as the dotted line 110, to obtain the continuous obstacle trajectory at the current time. As shown in
In some embodiments, the new trajectory may be connected to the continuous obstacle trajectory at the previous time. For example, from the previous time to the current time, the continuous obstacle may not have been interrupted, or the continuous obstacle may not have been blocked. In this case, the new trajectory, that is, the curve 150 and the continuous obstacle trajectory at the previous time, such as the dotted line 110, may be directly connected, as shown in
In some other embodiments, the new trajectory may not be connected to the continuous obstacle trajectory at the previous time. For example, from the previous time to the current time, the continuous obstacle may be interrupted, or the continuous obstacle may be blocked by other objects. The new trajectory obtained by clustering at the current time may not be directly connected with the continuous obstacle trajectory at the previous time, as shown in
In other embodiments, associating the cluster points in the cluster at the current time with the continuous obstacle trajectory at the previous time to obtain the continuous obstacle trajectory at the current time may include associating the cluster points in the cluster at the current time with the previous trajectory points in the continuous obstacle trajectory at the previous time to obtain the target points of the continuous obstacle trajectory at the current time; and, determining the continuous obstacle trajectory at the current time based on the target points of the continuous obstacle trajectory at the current time.
As shown in
In addition, in some embodiments, after clustering the stationary target points detected by the radar at the current time, the cluster obtained may not match the continuous obstacle trajectory at the previous time, and the quality of the cluster may be greater than the predetermined quality threshold. As shown in
In other embodiments, if the continuous obstacle trajectory at the previous time cannot be associated with the new trajectory corresponding to the cluster at each of the plurality of times after the previous time, then the part of the continuous obstacle trajectory at the previous time that exceeds the radar detection range will gradually disappear.
As shown in
In this embodiment, clustering processing can be performed on the stationary target points detected by the radar at the current time to obtain the cluster at the current time, and the degree of match between the cluster points at the current time and the continuous obstacle trajectory at the previous time can be calculated. If the degree of match between the cluster points at the current time and the continuous obstacle trajectory at the previous time is greater than the predetermined degree of match, the cluster points in the cluster at the current time can be associated with the continuous obstacle trajectory at the previous time to obtain the continuous obstacle trajectory at the current time. As such, the continuous obstacle trajectory can be continuously updated while the vehicle is moving.
An embodiment of the present disclosure provides a continuous obstacle detection system.
The radar 193 may include at least an antenna, and the antenna can be used to receive echo signals. The memory 191 can store program codes. The processor 192 can be configured to execute the program codes stored in the memory. When executed by the processor, the program codes can cause the processor to obtain the echo signal at the current time, and generate the detection data at the current time based on the echo signal; determine the stationary target points detected by the radar at the current time based on the detection data at the current time and the vehicle information of the vehicle at the current time; and determine a continuous obstacle trajectory at the current time based on the stationary target points detected by the radar at the current time.
In some embodiments, after determining the continuous obstacle trajectory at the current time based on the stationary target points detected by the radar at the current time, the processor 192 can be further configured to determine the boundary of the lane where the vehicle is positioned at the current time based on the continuous obstacle trajectory at the current time.
In some embodiments, the detection data may include one or more of the energy of a target point detected by the radar, or the distance, speed, angle of the target point relative to the radar.
In some embodiments, the vehicle information may include one or more of the speed, steering, or yaw rate of the vehicle.
In some embodiments, when determining the stationary target points detected by the radar at the current time based on the detection data at the current time and the vehicle information of the vehicle at the current time, the processor 192 may be configured to determine the stationary target points detected by the radar at the current time based on the speed of the vehicle at the current time, and the distance and speed of the target points detected by the radar at the current time relative to the radar.
In some embodiments, when determining the stationary target points detected by the radar at the current time based on the speed of the vehicle at the current time, and the distance and speed of the target points detected by the radar at the current time relative to the radar, the processor 192 may be configured to, if the distance of the target point relative to the radar detected by the radar at the current time is greater than a predetermined distance, compare the speed of the target point relative to the radar with the speed of the vehicle at the current time; and, if the different between the speed of the target point relative to the radar and the speed of the vehicle at the current time is less than a first predetermined difference, determine the target point as the stationary target point detected by the radar at the current time.
In some embodiments, when determining the stationary target points detected by the radar at the current time based on the speed of the vehicle at the current time, and the distance and speed of the target points detected by the radar at the current time relative to the radar, the processor 192 may be configured to, if the distance of the target point detected by the radar at the current time relative to the radar is less than or equal to the predetermined distance, determine the equivalent ground speed of the target point based on the angle of the target point relative to the radar; and compare the equivalent ground speed of the target point with the speed of the vehicle at the current time; and, if the difference between the equivalent ground speed of the target point and the speed of the vehicle at the current is less than a second predetermined difference, determine the target point as the stationary target point detected by the radar at the current time.
In some embodiments, the equivalent ground speed of the target point may be the radial speed of the target point relative to the radar.
In some embodiments, when determining the continuous obstacle trajectory at the current time based on the stationary target points detected by the radar at the current time, the processor 192 may be configured to generate the continuous obstacle trajectory at the current time based on the stationary target points detected by the radar at the current time.
In some embodiments, when generating the continuous obstacle trajectory at the current time based on the stationary target points detected by the radar at the current time, the processor 192 may be configured to perform clustering processing on the stationary target points detected by the radar at the current time to obtain the cluster at the current time; and generate the continuous obstacle trajectory at the current time based on the cluster at the current time if the quality of the cluster at the current time is greater than the predetermined quality threshold.
In some embodiments, before generating the continuous obstacle trajectory at the current time based on the cluster at the current time, the processor 192 may be further configured to determine the quality of the cluster at the current time based on one or more of the number of cluster points in the cluster at the current time, the length of the cluster, or the degree of match between the cluster and the vehicle information of the vehicle.
In some embodiments, when determining the continuous obstacle trajectory at the current time based on the stationary target points detected by the radar at the current time, the processor 192 may be configured to, based on the stationary target points detected by the radar at the current time, update the continuous obstacle trajectory at a previous time to obtain the continuous obstacle trajectory at the current time.
In some embodiments, when updating the continuous obstacle trajectory at a previous time to obtain the continuous obstacle trajectory at the current time based on the stationary target points detected by the radar at the current time, the processor 192 may be configured to perform clustering processing on the stationary target points detected by the radar at the current time to obtain the cluster at the current time; calculate the degree of match between the cluster points in the cluster at the current time and the continuous obstacle trajectory at the previous time; and associate the cluster points in the cluster at the current time with the continuous obstacle trajectory at the previous time to obtain the continuous obstacle trajectory at the current time if the degree of match between the cluster points in the cluster at the current time and the continuous obstacle trajectory at the previous time is greater than the predetermined degree of match.
In some embodiments, before calculating the degree of match between the cluster points in the cluster at the current time and the continuous obstacle trajectory at the previous time, the processor 192 may be further configured to determine the quality of the cluster at the current time based on one or more of the number of cluster points in the cluster at the current time, the length of the cluster, or the degree of match between the cluster and the vehicle information of the vehicle. When calculating the degree of match between the cluster points in the cluster at the current time and the continuous obstacle trajectory at the previous time, the processor 192 may be configured to calculate the degree of match between the cluster points in the cluster at the current time and the continuous obstacle trajectory at the previous time if the quality of the cluster at the current time is greater than the predetermined quality threshold.
In some embodiments, when calculating the degree of match between the cluster points in the cluster at the current time and the continuous obstacle trajectory at the previous time, the processor 192 may be configured to calculate the degree of match between the cluster points in the cluster at the current time and the continuous obstacle trajectory at the previous time based on the distance of the cluster points in the cluster relative to the radar at the current time and the distance of the previous trajectory points in the continuous obstacle trajectory at the previous time relative to the radar; and/or, calculate the degree of match between the cluster points in the cluster at the current time and the continuous obstacle trajectory at the previous time based on the speed of the cluster points in the cluster relative to the radar at the current time and the speed of the previous trajectory points in the continuous obstacle trajectory at the previous time relative to the radar.
In some embodiments, when calculating the degree of match between the cluster points in the cluster at the current time and the continuous obstacle trajectory at the previous time, the processor 192 may be configured to perform parameter fitting on the cluster at the current time to obtain the parameter information of the new trajectory corresponding to the cluster at the current time; and calculate the degree of match between the cluster points at the current time and the continuous obstacle trajectory at the previous time based on the parameter information of the new trajectory corresponding to the cluster at the current time and the parameter of the continuous obstacle trajectory at the previous time.
In some embodiments, when the processor 192 performs parameter fitting on the cluster at the current time, it may use one or more of a polynomial fitting method to perform parameter fitting on the cluster at the current time, or a radius arc fitting method to perform parameter fitting on the cluster at the current time.
In some embodiments, when associating the cluster points in the cluster at the current time with the continuous obstacle trajectory at the previous time to obtain the continuous obstacle trajectory at the current time, the processor 192 may be configured to obtain the new trajectory corresponding to the cluster at the current time based on the cluster points in the cluster at the current time; and, associate the new trajectory corresponding to the cluster at the current time with the continuous obstacle trajectory at the previous time to obtain the continuous obstacle trajectory at the current time, the continuous obstacle trajectory at the current time including the new trajectory and the continuous obstacle trajectory at the previous time.
In some embodiments, the new trajectory may be connected to the continuous obstacle trajectory at the previous time.
In some embodiments, the new trajectory may not be connected to the continuous obstacle trajectory at the previous time.
In some embodiments, when associating the cluster points in the cluster at the current time with the continuous obstacle trajectory at the previous time to obtain the continuous obstacle trajectory at the current time, the processor 192 may be configured to associate the cluster points in the cluster at the current time with the previous trajectory points in the continuous obstacle trajectory at the previous time to obtain the target points of the continuous obstacle trajectory at the current time; and, determine the continuous obstacle trajectory at the current time based on the target points of the continuous obstacle trajectory at the current time.
In some embodiments, if the continuous obstacle trajectory at the previous time cannot be associated with the new trajectory corresponding to the cluster at each of the plurality of times after the previous time, then the part of the continuous obstacle trajectory at the previous time that exceeds the radar detection range will gradually disappear.
In some embodiments, the radar may be a millimeter wave radar.
In this embodiment, the echo signal at the current time can be obtained, the detection data at the current time can be generated based on the echo signal, the stationary target points detected by the radar at the current time can be determined based on the detection data at the current time and the vehicle information of the vehicle at the current time, and continuous obstacle trajectory at the current time can be determined based on the stationary target points detected by the radar at the current time, thereby realizing the detection of continuous obstacles around the vehicle.
An embodiment of the present disclosure provide a vehicle. The vehicle may include a vehicle body, a power system, and the continuous obstacle detection system described in the above embodiment. The power system can be disposed on the vehicle body to provide power. The implementation method and specific principles of the continuous obstacle detection system are consistent with the foregoing embodiment, which will not be repeated here.
In addition, an embodiment of the present disclosure provides a computer-readable storage medium having a computer program stored thereon. The computer program can be executed by a processor to implement the continuous obstacle detection method described in the foregoing embodiments.
In the several embodiments provided by the present disclosure, it should be understood that the disclosed apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative. For example, the unit division is merely logical function division and there may be other division in actual implementation. For example, multiple units or components may be combined or integrated into another system, or some features can be omitted or not be executed. In addition, the mutual coupling or the direct coupling or the communication connection as shown or discussed may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical, mechanical or other forms.
The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, may be located in one place or may also be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solution in the disclosure.
In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of hardware plus software functional unit.
The above-described integrated unit implemented in the form of a software functional unit may be stored in a computer-readable storage medium. The software function unit is stored in a storage medium and includes several instructions for enabling a computer device (which may be a personal computer, a server, a network device, etc.) or a processor to execute some steps of the method according to each embodiment of the present disclosure. The foregoing storage medium includes a medium capable of storing program code, such as a USB flash disk, a removable hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, an optical disc, or the like.
Those skilled in the art may clearly understand that, for convenience and brevity of description, the division of the foregoing functional modules is only used as an example. In practical applications, however, the above function allocation may be performed by different functional modules according to actual needs. That is, the internal structure of the device is divided into different functional modules to accomplish all or part of the functions described above. For the working process of the foregoing apparatus, reference may be made to the corresponding process in the foregoing method embodiments, and details are not described herein again.
Finally, it should be noted that the foregoing embodiments are merely intended for describing the technical solutions of the present disclosure, but not to limit the present disclosure. Although the present disclosure is described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that the technical solutions described in the foregoing embodiments may still be modified, or a part or all of the technical features may be equivalently replaced without departing from the spirit and scope of the present disclosure. As a result, these modifications or replacements do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the present disclosure.
This application is a continuation application of International Application No. PCT/CN2018/124887, filed on Dec. 28, 2018, the entire content of which is incorporated herein by reference.
| Number | Date | Country | |
|---|---|---|---|
| Parent | PCT/CN2018/124887 | Dec 2018 | US |
| Child | 17202322 | US |
