Patent Grant
11136223
Within many industrial facilities, objects are transported on conveyor belts from one location to another. Often a conveyor belt will carry an unsorted mixture of various objects and materials. In some instances, like within recycling and waste management facilities for example, some of the objects may be considered desirable (e.g. valuable) materials while others may be considered undesirable contaminants. For example, the random and unsorted contents of a collection truck may be unloaded at the facility onto a conveyor belt. Although sorting personnel may be stationed to manually sort materials as it is transported on the belt, the use of sorting personnel is limiting because they can vary in their speed, accuracy and efficiency and can suffer from fatigue over the period of a shift. Human sorters also require specific working conditions, compensation, and belt speeds. Production time is lost to training the many new employees that enter as sorters, and operation costs increase as injuries and accidents occur.
The introduction of recycling robotics for sorting materials has led to increased productivity and decreased contamination for Material Recovery Facilities (MRFs). Robots have been utilized as a viable replacement, or supplement, for human sorters due to their speed, reliability, and durability. The objective of recycling robots is to recover the specific target material(s) and eject them into bunkers without introducing other materials (contaminants) into the sorted bunkers. A common technique used by these recycling robots involves the use of suction gripper. A suction cup gripper connected to a pneumatic system would generate a substantial suction force to grasp targeted objects. Application of the suction force may be curtailed once the objected is picked from conveyor belt to direct the item into the proper collection bunker.
Robotic sorting has proven to be exceptionally good at picking and placing rigid objects with surface areas greater than or equal to three square inches. Non-rigid objects such as plastic bags have a lower success rate due to their tendency to deform and lose viable surface area to wrinkles and creases and interfere with suction. Objects smaller than three square inches in cross sectional area have trouble being picked and placed into appropriate bins since suction cups have difficulty sealing on them. Moreover, the robotic apparatus used to position the suction gripper, while less expensive over time as compared to human sorters, still require a significant capital expense that leaves some material economically infeasible for collection and recycling.
For example, many Material Recovery Facilities request their customers not include items smaller than three square inches and thin film in their recycling, even though they are made of recyclable materials. Material Recovery Facilities will often tend to avoid retrieving paper since an excessive number of material and physical picks are required to accumulate an economical amount of mass. Rapidly moving mechanical elements can also present a hazard to facility personnel working near the sorting robot.
For the reasons stated above and for other reasons stated below which will become apparent to those skilled in the art upon reading and understanding the specification, there is a need in the art for system and methods for vacuum extraction for material sorting applications.
In one embodiment, a vacuum object sorting system comprises: a vacuum extraction assembly that includes at least one vacuum extractor device having an inlet and an outlet, wherein the at least one vacuum extractor device is configured to convert a controlled compressed air stream into a channeled vacuum airflow entering the inlet and exiting the outlet; and an object recognition device coupled to sorting control logic and electronics, wherein the controlled compressed air stream is controlled by the sorting control logic and electronics in response to a signal generated by the object recognition device; wherein the at least one vacuum extractor device is configured to capture a target object identified by the sorting control logic and electronics utilizing the channeled vacuum airflow, and further utilizing the channeled vacuum airflow, to pass the target object through the inlet and outlet to a deposit location.
Embodiments of the present disclosure can be more easily understood and further advantages and uses thereof more readily apparent, when considered in view of the description of the preferred embodiments and the following figures in which:
In accordance with common practice, the various described features are not drawn to scale but are drawn to emphasize features relevant to the present disclosure. Reference characters denote like elements throughout figures and text.
In the following detailed description, reference is made to the accompanying drawings that form a part hereof, and in which is shown by way of specific illustrative embodiments in which the embodiments may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the embodiments, and it is to be understood that other embodiments may be utilized and that logical, mechanical and electrical changes may be made without departing from the scope of the present disclosure. The following detailed description is, therefore, not to be taken in a limiting sense.
The present disclosure illustrates various example embodiments of vacuum extraction for material sorting applications. As discussed below, a sorting machine based on vacuum extraction as described herein can quickly and accurately remove materials from a moving conveyor mechanism to allow facility operators, such as but not limited to Material Recovery Facilities, to consider a wider variety of recyclable materials and/or remove non rigid contaminants in an efficient and effective manner. In this way, smaller rigid and non-rigid materials may be captured along with relatively larger rigid materials using vacuum extraction as described herein.
More specifically, several of the embodiments presented herein disclose a vacuumed extraction assembly. In some embodiments, the vacuum extraction assembly, which would take the place of the robotic elements and suction gripper of known systems, may be (but is not limited to) a mechanically static apparatus that removes objects from a moving conveyor line (for example) using a plurality of individual vacuum extractors positioned over the moving conveyor. In some embodiments, an external control system and visual recognition system will determine which vacuum extractor of the vacuum extraction assembly to engage to pick a target object and send control signals to execute the capture action.
In the example shown in
In some implementations, vacuum sorting system 10 further comprises at least one object recognition device 162. In some embodiments, the object recognition device may comprise an imaging device 162 (such as, for example, an infrared camera, visual spectrum camera, or a some combination thereof) directed at a conveyer belt 50 that transports target objects (shown at 55) within the operating reach of the vacuum sorting system 10. In some embodiments, the imaging device 162 produces a signal that is delivered to the sorting control logic and electronics 160 and which may be used by sorting control logic and electronics 160 to send airflow control signals to the vacuum control system 140 of the vacuum sorting system 10 (and in some embodiment the vacuum extraction assembly 100), in order to initiate a capture action on a target object 55. It should be understood that utilizing an imaging device for the object recognition device 162 is presented as an example implementation. The embodiments described herein, however, may implement the object recognition device 162 utilizing any form of a sensor configured for detecting, for example, non-visible electromagnetic radiation (such as a hyperspectral camera, infrared, or ultraviolet), a magnetic sensor; a capacitive sensor; or other sensors commonly used in the field of industrial automation. As such, the signal that is delivered to the sorting control logic and electronics 160 from the object recognition device 162 may comprise, but is not necessarily, an image signal.
As shown in
In various embodiments, the sorting control logic and electronics 160 may be implemented using a microprocessor coupled to a memory that is programed to execute code to carry out the functions of the sorting control logic and electronics 160 described herein. In other embodiments, the sorting control logic and electronics 160 may additionally, or alternately, be implemented using an application specific integrated circuit (ASIC) or field programmable gate array (FPGA) that has been adapted for machine learning. In operation, in some embodiments, the imaging device 162 is directed towards the conveyor belt 50 in order to capture an overhead view of the materials 55 being transported by the conveyor belt 50. The imaging device 162 produces a signal that is delivered to the sorting control logic and electronics 160.
In some embodiments, within the sorting control logic and electronics 160, these image frames are provided input to one or more neural network and artificial intelligence algorithms (shown as the Neural Processing Units 164) to locate and identify material appearing within the image frames. A feed of image frames captured by the imaging device 162 is fed to a machine learning inference algorithm implemented by the Neural Processing Units 164. The sequence of captured image frames may be processed by multiple processing layers, or neurons, of the Neural Processing Units 164 to evaluate the correlation of specific features with features of objects that it has previously learned. Alternative algorithms to detect objects within an image include Fully Convolutional Neural Network, Multibox, Region-based Fully Convolutional Networks (R-FCN), Faster R-CNN, and other techniques commonly known to those skilled in the art as object detection, instance-aware segmentation, or semantic segmentation algorithms described in available literature. Further details regarding examples of the process of detecting objects in captured images which may be used in the present embodiments may be found in the reference Huang et al., “Speed/accuracy trade-offs for modern convolutional object detectors” The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, pp. 7310-7311, which is incorporated by reference in its entirety.
The vacuum control system 140 provides an air supply for operating and individually controlling the vacuum extractor devices 110 of the vacuum extraction assembly 100. When the object recognition device 162 identifies an object 55 to remove from the conveyor belt 55, an electrical signal is sent to the vacuum control system 140 to activate an air supply. As illustrated in
While the vacuum extractor device 110 may be controlled by the sorting control logic and electronics 160 in response to signals from the object recognition device 162, it should be understood that in other embodiments, a vacuum extractor device 110 may be manually controlled by an operator (either locally or remotely). Alternatively, they may be responsive to either manual controls and/or control signals from the sorting control logic and electronics 160. For example, the sorting control logic and electronics 160 may be bypassed by a manual override that permits the operator to control individual, or groups of, vacuum extractor device 110.
In some embodiments, each vacuum extractor device 110 comprises a vacuum generator 220 that generates the vacuum force used to extract the object 55 from the conveyor 50, as opposed to merely communicating a vacuum force generated by the vacuum control system 140. In some embodiments, vacuum generator 220 is coupled to the vacuum control system 140 by air supply tubing 225. In some embodiments, the vacuum generator 220 comprises a compressed air driven vacuum generator, such as a Venturi and/or Coanda vacuum generator for example. That is the motive force that pulls a vacuum through vacuum extractor device 110 is the result of a flow of a compressed air stream, supplied by an air source 145 of vacuum control system 140 that flows through the vacuum generator 220. In the example embodiments depicted in
In operation, the positive pressure airflow applied at a pressurized air input port 225 flows across one or more Venturi and/or Coanda feature within vacuums generators 220 to create a negative air pressure that pulls air into the intake of the vacuum extraction device 110, thus creating suction at the intake capable of pulling objects 55 in and through the vacuum extraction device 110. One example of a device comprising such a Venturi vacuum generator is the “Adjustable Inducer/Venture System” by Airtrim Pneumatic Conveyance Systems described in U.S. Pat. No. 6,732,897, the contents of which are incorporated by reference herein in their entirety. See also U.S. Pat. No. 3,716,307 and U.S. Patent Publication 2004/0055250, the contents of both of which are incorporated by reference herein in their entirety.
Once the target object 55 is removed from the conveyor belt 50 and passes through the vacuum extraction device 110, in some embodiments, the materials may be transported by a hood, hoses, ducts or tubes 130 leading to a holding bin, tank, bunker or other receptacle 135 where extracted objects 55 are deposited. The particular destination for items removed from the conveyor belt may depend upon whether they are contaminants or desired materials. In some embodiments the receptacle 135 may be adjacent to the vacuum sorting system 10 while in others, it may be remotely located away from the vacuum sorting system 10. In some embodiments, the receptacle 135 may comprise a cargo area of a truck or other vehicle so that removed objects 55 are directly loaded onto the vehicle for transport. In some embodiments, the hood, hoses, ducts or tubes 130 may include controllable valves or other controllable diverters that control the material flow of removed objects 55 that have entered the suction ducting 130 in order that various object disposal locations (that is, multiple alternate receptacles 135) may be selected for any of the plurality of the vacuum extractor device 110. That is, the ducting 130 may be configurable and re-configurable using the controllable valves or other controllable diverters (by the electronics 160 or other controller) such that objects 55 extracted by one vacuum extractor device 110 of assembly 100 may be routed to a different receptacle 135 than objects 55 extracted by another one vacuum extractor device 110 of assembly 100. Moreover, if a receptacle 135 is reaching full capacity, the ducting 130 may be re-configured to route extracted objects 55 to a different receptacle 135. For example,
In some embodiments, vacuum control system 140 utilizes an air manifold 143 or some other pressurized air distribution mechanism. In some embodiments, the air manifold 143 may be further pneumatically coupled to the air source 145. In alternate implementations, the air source 145 may comprise a blower, an air compressor, a compressed air storage tank, or some combination thereof. Although this disclosure may refer to “air” with regards to “airflow”, “air compressor” and other elements, it should be understood that the term “air” is used in a generic sense to refer to any compressible gas or mixture of gasses.
The air manifold 143 may comprise a series of control valves 147 that may operate in response to control signal generated by the sorting control logic and electronics 160. As such, to communicate control signals, sorting control logic and electronics 160 may further comprise elements to generate electrical and/or control pneumatic signals to the vacuum extraction assembly 100. In some embodiments, the control signals may be used to turn the suction force applied by a vacuum extractor device 110 on or off. In some embodiments, the control signals may adjust a control valve 147 to vary air flow applied to a vacuum extractor device 110 and thus regulate a vacuum force applied by that vacuum extractor device 110 on a target object 55. In still other embodiments, the vacuum extractor device 110 may be optionally configured to use the air flow from the air manifold 143 to produce a reverse airflow through the vacuum extractor device 110, for example, in order to blow out material that may have become lodged in the intake of the vacuum extractor device 110. In some embodiments, reversal of airflow through vacuum extractor device 110 may be controlled by the sorting control logic and electronics 160.
In some embodiments, each vacuum extractor device 110 may use a sensor (for example, at their outlet) to send feedback to the sorting control logic and electronics 160 when a collected item fully passes through the vacuum extractor device 110, or to indicate when a collected item has not fully passed through the vacuum extractor device 110 (for example, become lodged). This will allow the sorting control logic and electronics 160 to operate the air manifold 143, valves 147 and or vacuum extractor device 110 to turn off the vacuum force off at the appropriate time (and conserve energy) and/or reverse the airflow through a vacuum extraction device 110 if necessary. Similarly, the sensor output may be used to detect a clog or jam and trigger raising or lowering of the vacuum extraction assembly 100. Such a sensor is not limited to any particular technology, and may comprise, for example, a pressure sensor, ultrasonic sensor, infrared sensor, opacity sensor, or the like.
It should be understood that in alternate implementations, the vacuum extractor devices 110 may be positioned on the vacuum extraction assembly 100 in various arrangements or geometries. That is, in some embodiments, the vacuum extraction assembly 100 may comprise a single row of vacuum extraction devices 110 arranged in a line across the conveyor device 50 perpendicular with respect to the direction of material travel. In other embodiments, such as shown in
Where the vacuum extraction assembly 100 does comprise a plurality of vacuum extraction devices 110, they need not be uniform in size. For example, a vacuum extraction assembly 100 may comprise one or more vacuum extraction devices 110 of a first size, and one or more vacuum extraction devices 110 of a second size. They also need not be uniform in geometry. For example, the sorting control logic and electronics 160 may determine that the a target object 55 has a certain characteristic (for example, size, shape, orientation, material type or composition or any other characteristic discernable by the sorting control logic and electronics 160) and correlated that characteristic with a specific vacuum extraction device 110 of the vacuum extraction assembly 100 best suited for capturing objects having that characteristic. Vacuum extraction device 110 with wider diameters may be used for flexible materials like bags and sheets with smaller diameters best for more ridged objects. For example, an object identified as being a disposable ground-coffee pod may be selected for extraction by a first vacuum extraction device 110 of a first size, while a sheet of plastic wrap may be selected for extraction by a first vacuum extraction device 110 of a second size. In some embodiments, the Neural Processing Units 164 outputs one or more physical object attributes determined by the one or more Neural Processing Units based on the visional inspection of the one or more target objects appearing in captured image frames.
Because operation of the vacuum extraction assembly 100, in some embodiments, may be substantially motionless, the cost of installation and maintenance will remain substantially lower than dynamic robots. Blockage in the vacuum extraction devices 110 can more easily be removed, allowing reliable and continuous removal of target object 55 from the conveyor 50. In some embodiments, adjustments can be made on the conveyor 50 to increase or focus flow, including but not limited to adding perforations on the belt, replacing the belt with a material that allows air to flow through it, or gaps in the conveyor beneath the vacuums. The control air may be optimized by adding sensors (for example, to the inlet or outlet of the vacuum extractor devices 110) to determine when to disable flow or adding air pressure amplifiers to the system. In some embodiments, the addition of accessories (such as at 210) at the inlet of the vacuum extractor devices 100, such as but not limited to longer, spiraled, frictionless, or wider funnels, sorters or shredders add versatility, allowing adjustments to accommodate target objects 55 of many shapes, sizes, rigidity and densities. This could make the removal of materials smaller in size—such as bottle caps—or lower in density—such as paper or plastic film—economically viable. Using the vacuum extraction assembly 100 will therefor allow material recovery facilities to expend more effort into collecting hard-to-recycle items (for example, like those with less than three square inches of surface area or non-rigid materials such as plastic film or paper). It will also help increase the worth of recovered material by removing contaminants such as thin film from the stream.
It should be understood that components, elements and features of any of the embodiments described herein may be used in combination. Moreover, it should be understood that in some embodiments, vacuum sorting system 10 may be used in combination or conjunction with robotic sorting systems such as those comprising suction grippers. As such, other embodiments are intended to include sorting systems that may comprise both suctions grippers and a vacuum extraction assembly as described herein.
In addition to the embodiments described above, additional features may include: the vacuum extraction assembly 100 comprising a series of inlets configured to produce a channeled vacuum flow placed over a conveyor, for use in removing objects from the conveyor and conveying them to a deposit location, where the choice of enabling suction for object removal is governed by a computer vision system. A vacuum extraction assembly 100 wherein the vacuum flow is created by a straight-through Venturi or Coanda vacuum generator that utilizes compressed air for the generation of the vacuum and is placed inline in the vacuum channel. A vacuum extraction assembly 100 wherein the vacuum flow is enabled or disabled through the use of a valve controlling compressed air to the vacuum generator. A vacuum extraction assembly 100 wherein the object recognition device 162 is configured for detecting non-visible electromagnetic radiation, such as a hyperspectral camera, infrared, or ultraviolet sensor; a magnetic sensor; a capacitive sensor; or other sensors commonly used in the field of industrial automation. A vacuum extraction assembly 100 wherein the sorting control logic and electronics 160 is one that uses a set of parameters for identification on a series of color photos provided by a camera, and where the parameters have been trained using machine learning. A vacuum extraction assembly 100 wherein a vacuum extractor 110 inlet has an attachment to channel flow. A vacuum extraction assembly 100 used in conjunction with a conveyor mechanism 50 or conveyor belt 50 that comprises holes 51. A vacuum extraction assembly 100 wherein the vacuum is generated by a blower, and the vacuum is disabled by flooding the vacuum inlet with compressed air. A vacuum extraction assembly 100 wherein the vacuum outlet has a sensor to detect the passing of target materials and determine when the airflow must be disabled. A vacuum extraction assembly 100 wherein the source of air is amplified to produce higher inlet pressures thus higher flow rates. A vacuum extraction assembly 100 wherein the conveyor has a lower coefficient of friction due to airflow through a porous conveyor belt. A vacuum extraction assembly 100 with inlets comprising an accessory with a specific angle to guide larger but non-rigid material through the vacuum and prevent blockage upon entry. A vacuum extraction assembly 100 wherein the hoses attached to a Venturi outlet channel objects to a separate object disposal location. A vacuum extraction assembly 100 wherein the inlet and exhaust of a vacuum generator are augmented by fitting to improve the flow through the use of a converging cone on the inlet, and a diverging cone on the exhaust. A vacuum extraction assembly 100 wherein multiple units of an object recognition device 162 are placed between individual vacuum extraction assemblies to provide feedback as target objects move. A vacuum extraction assembly wherein the vacuum extraction assemblies are positioned to follow a mechanism that singulates the stream of materials and guarantees a single layer of objects for higher recovery. A vacuum extraction assembly wherein there are attachments at the inlets to allow separation of target materials from items that may be attached to it at the time of recovery. In some embodiments, the attachments comprising teeth configured to shred materials entering a vacuum extracting device. In some embodiments, the attachments comprising sorting features configured to separate target objects and destroy undesired materials.
The method 500 begins at 510 with controlling a compressed air stream using sorting control logic and electronics in response to a signal generated by an object recognition device. As described above, the object recognition device may be any form of sensing apparatus for detecting a target object for extraction, and may for example comprise a camera, an infrared camera, a non-visible electromagnetic radiation sensor, a magnetic sensor or a capacitive sensor, or some combination of such sensors. The sensing apparatus captures one or more characteristics about the target object, including but not limited to the material, location, relation to nearby objects, and other features.
As shown at 520, the method 500 further includes converting the controlled compressed air stream into a channeled vacuum airflow using a vacuum extraction assembly that includes one or more vacuum extractor devices each having an inlet and an outlet, wherein the channeled vacuum airflow enters the inlet and exits the outlet. In some embodiments, the vacuum extraction assembly may comprise a plurality of individual vacuum extractor devices. In some embodiments, the vacuum extraction assembly comprises a plurality of vacuum extractor devices individually operated by the sorting control logic and electronics via a respective controlled compressed air stream.
For example, the using sorting control logic and electronics can select which of the vacuum extractor devices to activate based upon the location of the target object, the material composition of the object, and other factors or characteristics of the object as detected or identified by the sorting control logic and electronics. The compressed air stream associated with the selected vacuum extractor device(s) is then controlled in response to a signal generated by the control logic and electronics. The signal to actuate the capture action may controlled in part by timing logic—time-based, mechanical, or otherwise—that determines the optimal time to attempt to extract the object.
In some embodiments, a conveyor mechanism may be used to channel the target object past the object recognition device and towards the vacuum extraction assembly. Such a conveyor mechanism may comprise, but is not necessarily limited to, a conveyor belt. In some embodiments, the conveyor mechanism may comprise one or more raised features, recessed features, or perforation features, configured to provide airflow for intake into the channeled vacuum airflow (for example, holes, cleats, treads, or surface textures).
The method 500 proceeds to 530 with utilizing the channeled vacuum airflow to capture a target object identified by the sorting control logic and electronics by drawing the target object in through the inlet of a first vacuum extractor device of the one or more vacuum extractor devices and out through the outlet of the first vacuum extractor device. The method 500 proceed then to 540 with discharging the target object to a deposit location. The deposit location may comprise any form of holding bin, tank, bunker, vehicle, or other receptacle where extracted objects are deposited. In some embodiments, the method 500 may optionally further proceed to 550 with determining when the target object passes through the outlet of the first vacuum extractor device using a sensor. The sorting control logic and electronics may then take further actions in response to a signal from that sensor. For example, failure to sense a successful passage our from the outlet of a vacuum extractor device may be an indication that the capture action failed (e.g., that the identified target object was missed or that a sufficient airflow to lift the target object into the inlet was not applied). Alternatively failure to sense a successful passage our from the outlet of a vacuum extractor device may be an indication that the that vacuum extractor device has become jammed or clogged. In this case, the sorting control logic and electronics may respond by reversing the airflow through the vacuum extractor device as described above to eject the clogged material from the inlet. These examples are not intended to be limiting. In other embodiments, other action may be performed by the logic in response to the detection of an obstruction.
Example 1 includes a vacuum object sorting system, the system comprising: a vacuum extraction assembly that includes at least one vacuum extractor device having an inlet and an outlet, wherein the at least one vacuum extractor device is configured to convert a controlled compressed air stream into a channeled vacuum airflow entering the inlet and exiting the outlet; and an object recognition device coupled to sorting control logic and electronics, wherein the controlled compressed air stream is controlled by the sorting control logic and electronics in response to an signal generated by the object recognition device; wherein the at least one vacuum extractor device is configured to capture a target object identified by the sorting control logic and electronics utilizing the channeled vacuum airflow, and further utilizing the channeled vacuum airflow, to pass the target object through the inlet and outlet to a deposit location.
Example 2 includes the system of example 1, wherein the vacuum extraction assembly comprises a plurality of vacuum extractor devices individually operated by the sorting control logic and electronics via a respective controlled compressed air stream.
Example 3 includes the system of example 2, wherein a first vacuum extractor device of the plurality of vacuum extractor devices is selected for operation to capture the target object based on a characteristic of the target objection identified by the sorting control logic and electronics.
Example 4 includes the system of example 3, wherein the characteristic comprises one or more of a size, a shape, an orientation, a material type, or a material composition of the target object.
Example 5 includes the system of any of examples 1-4, wherein the at least one vacuum extractor device comprises a sensor that detects when the target object passes through the at least one vacuum extractor device.
Example 6 includes the system of example 5, wherein the sensor sends a feedback signal to the sorting control logic and electronics; wherein the sorting control logic and electronics controls the controlled compressed air stream in response to the feedback signal.
Example 7 includes the system of any of examples 5-6, wherein the sorting control logic and electronics controls detect an obstruction based on the feedback signal.
Example 8 includes the system of any of examples 1-7, wherein the object recognition device comprises at least one of: a camera, an infrared camera, a non-visible electromagnetic radiation sensor, a magnetic sensor or a capacitive sensor.
Example 9 includes the system of any of examples 1-8, further comprising a conveyor mechanism configured to channel the target object past the object recognition device and towards the vacuum extraction assembly.
Example 10 includes the system of example 9, wherein the conveyor mechanism comprises one or more raised features, recessed features, or perforation features, configured to provide airflow for intake into the channeled vacuum airflow.
Example 11 includes the system of any of examples 1-10, further comprising a ducting system coupled to the outlet of the at least one vacuum extractor device, further comprising a ducting system coupled to the outlet of the at least one vacuum extractor device, wherein the ducting system is reconfigurable to direct individual target objects to one of a plurality of different receptacles.
Example 12 includes the system of any of examples 1-11, wherein the sorting control logic and electronics comprises one or more neural processing units, wherein the one or more neural processing units distinguish the target object from one or more non-target based on a neural network parameter set.
Example 13 includes the system of example 12, wherein the one or more neural processing units further identify predefined characteristics of the target object based on the neural network parameter set.
Example 14 includes a method for vacuum object sorting, the method comprising: controlling a compressed air stream using sorting control logic and electronics in response to a signal generated by an object recognition device; controlling the compressed air stream using sorting converting the controlled compressed air stream into a channeled vacuum airflow using a vacuum extraction assembly that includes one or more vacuum extractor devices each having an inlet and an outlet, wherein the channeled vacuum airflow enters the inlet and exits the outlet; utilizing the channeled vacuum airflow to capture a target object identified by the sorting control logic and electronics by drawing the target object in through the inlet of a first vacuum extractor device of the one or more vacuum extractor devices and out through the outlet of the first vacuum extractor device; and discharging the target object to a deposit location.
Example 15 includes the method of example 14, wherein the vacuum extraction assembly comprises a plurality of vacuum extractor devices individually operated by the sorting control logic and electronics via a respective controlled compressed air stream.
Example 16 includes the method of example 15, further comprising: selecting the first vacuum extractor device for operation to capture the target object based on a characteristic of the target objection identified by the sorting control logic and electronics.
Example 17 includes the method of any of examples 14-16, wherein the first vacuum extractor device comprises a sensor that detects when the target object passes through the at least one vacuum extractor device and sends a feedback signal to the sorting control logic and electronics, the method further comprising: controlling the controlled compressed air stream in response to the feedback.
Example 18 includes the method of any of examples 14-17, wherein the object recognition device comprises at least one of: a camera, an infrared camera, a non-visible electromagnetic radiation sensor, a magnetic sensor or a capacitive sensor.
Example 19 includes the method of any of examples 14-18, further comprising: channeling the target object past the object recognition device and towards the vacuum extraction assembly utilizing a conveyor mechanism.
Example 20 includes the method of example 19, wherein the conveyor mechanism comprises one or more raised features, recessed features, or perforation features, configured to provide airflow for intake into the channeled vacuum airflow.
In various alternative embodiments, system elements, method steps, or examples described throughout this disclosure (such as the sorting control logic and electronics, vacuum control system, neural processing units and/or sub-parts of any thereof, for example) may be implemented using one or more computer systems, field programmable gate arrays (FPGAs), or similar devices and/or comprising a processor coupled to a memory and executing code to realize those elements, processes, steps or examples, said code stored on a non-transient data storage device. Therefore other embodiments of the present disclosure may include elements comprising program instructions resident on computer readable media which when implemented by such computer systems, enable them to implement the embodiments described herein. As used herein, the term “computer readable media” refers to tangible memory storage devices having non-transient physical forms. Such non-transient physical forms may include computer memory devices, such as but not limited to punch cards, magnetic disk or tape, any optical data storage system, flash read only memory (ROM), non-volatile ROM, programmable ROM (PROM), erasable-programmable ROM (E-PROM), random access memory (RAM), or any other form of permanent, semi-permanent, or temporary memory storage system or device having a physical, tangible form. Program instructions include, but are not limited to computer-executable instructions executed by computer system processors and hardware description languages such as Very High Speed Integrated Circuit (VHSIC) Hardware Description Language (VHDL).
As used herein, terms such as “sorting control logic and electronics”, “vacuum control system”, “vacuum extraction assembly”, “vacuum extractor device”, “imaging device”, “neural processing units”, “data storage”, “memory device”, “memory”, “neural network parameter set”, “controller”, “controls”, “processor”, or “microprocessor”, each refer to non-generic device elements that would be recognized and understood by those of skill in the art and are not used herein as nonce words or nonce terms for the purpose of invoking 35 U.S.C. 112(f).
Although specific embodiments have been illustrated and described herein, it will be appreciated by those of ordinary skill in the art that any arrangement, which is calculated to achieve the same purpose, may be substituted for the specific embodiment shown. This application is intended to cover any adaptations or variations of the presented embodiments. Therefore, it is manifestly intended that embodiments be limited only by the claims and the equivalents thereof.
This patent application is a U.S. patent application claiming priority to, and the benefit of, U.S. Provisional Patent Application No. 62/732,689, titled “VACUUM EXTRACTION FOR MATERIAL SORTING APPLICATIONS” filed Sep. 18, 2018, which is incorporated herein by reference in its entirety.
| Number | Name | Date | Kind |
|---|---|---|---|
| 3716307 | Hansen | Feb 1973 | A |
| 5040353 | Evans | Aug 1991 | A |
| 5864600 | Gray et al. | Jan 1999 | A |
| 5956413 | Oste et al. | Sep 1999 | A |
| 6124560 | Roos | Sep 2000 | A |
| 6732897 | DeHart et al. | May 2004 | B2 |
| 7516848 | Shakes et al. | Apr 2009 | B1 |
| 9785851 | Torek et al. | Oct 2017 | B1 |
| 10016790 | Travis | Jul 2018 | B1 |
| 10298909 | Peeters | May 2019 | B2 |
| 10799915 | Horowitz | Oct 2020 | B2 |
| 20040055250 | Main et al. | Mar 2004 | A1 |
| 20100131094 | Green | May 2010 | A1 |
| 20120293623 | Nygaard | Nov 2012 | A1 |
| 20180243800 | Kumar | Aug 2018 | A1 |
| 20190374978 | Borrell | Dec 2019 | A1 |
| Number | Date | Country |
|---|---|---|
| 2008150903 | Dec 2008 | WO |
| 2011115481 | Sep 2011 | WO |
| 2012067512 | May 2012 | WO |
| Entry |
|---|
| International Searching Authority, “Invitation to Pay Additional Fees from PCT Application No. PCT/US2019/049361”, from Foreign Counterpart to U.S. Appl. No. 16/559,299, dated Nov. 20, 2019, pp. 1-12, Published: WO. |
| Huang et al., “Speed/accuracy trade-offs for modern convolutional object detectors”, IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Jul. 2017, pp. 7310-7319, IEEE. |
| International Searching Authority, “International Search Report and Written Opinion from PCT Application No. PCT/US2019/049361”, from Foreign Counterpart to U.S. Appl. No. 16/559,299, dated Feb. 20, 2020, pp. 1-19, Published: WO. |
| Number | Date | Country | |
|---|---|---|---|
| 20200087118 A1 | Mar 2020 | US |
| Number | Date | Country | |
|---|---|---|---|
| 62732689 | Sep 2018 | US |
