Object path planning in a sorting facility

Patent metadata

Field	Value
Publication number	US-12221292-B2
Application number	US-202117559798-A
Country	US
Kind code	B2
Filing date	Dec 22, 2021
Priority date	Dec 22, 2021
Publication date	Feb 11, 2025
Grant date	Feb 11, 2025

Abstract

Object path planning in a sorting facility is disclosed, including: obtain data describing a trajectory associated with a target object; generate a control signal for a sorting device to perform a sorting operation on the target object based at least in part on the trajectory associated with the target object; and provide the control signal to the sorting device, wherein the sorting device is configured to execute the control signal with respect to the target object.

First claim

What is claimed is:

1. A system, comprising:
a processor configured to:
obtain a plurality of image frames showing a target object;
determine a set of velocities associated with a movement of the target object based at least in part on the plurality of image frames and time intervals between image frames, wherein the set of velocities corresponds to a set of dimensions associated with a coordinate system;
generate a trajectory associated with the target object based at least in part on the set of velocities associated with the movement of the target object;
generate a control signal for a sorting device to perform a sorting operation on the target object based at least in part on the trajectory associated with the target object; and
provide the control signal to the sorting device, wherein the sorting device is configured to execute the control signal with respect to the target object; and

a memory coupled to the processor and configured to provide the processor with instructions.

2. The system of claim 1 , wherein the processor is further configured to publish an updated trajectory associated with the target object synchronous with one or more captures of new sets of sensed data associated with the target object.

3. The system of claim 1 , wherein the processor is further configured to publish an updated trajectory associated with the target object asynchronous to captures of new sets of sensed data associated with the target object.

4. The system of claim 1 , wherein the control signal instructs the sorting device to omit performing the sorting operation on the target object.

5. The system of claim 1 , wherein to generate the control signal for the sorting device is based at least in part on a location associated with the sorting device, a location associated with the target object, or both.

6. The system of claim 1 , wherein to generate the control signal for the sorting device to perform the sorting operation on the target object based at least in part on the trajectory associated with the target object comprises to include in the control signal a sorting parameter associated with the sorting operation on the target object, wherein the sorting parameter is determined based at least in part on the trajectory.

7. The system of claim 6 , wherein the sorting parameter comprises a specified depth at which to perform the sorting operation.

8. The system of claim 6 , wherein the sorting parameter comprises a specified force at which to perform the sorting operation.

9. The system of claim 6 , wherein the sorting parameter is further determined by an object type associated with the target object.

10. The system of claim 1 , where sensed data is input to a machine learning model to identify the target object.

11. The system of claim 1 , wherein the control signal specifies a location at which the sorting device is to perform the sorting operation on the target object.

12. The system of claim 1 , wherein the processor is further configured to update the trajectory of the target object based at least in part on new sensed data associated with the target object.

13. The system of claim 1 , wherein the system further includes a sensor, wherein the processor is configured to:
receive, from the sensor, sensed data associated with the target object; and
apply a machine learning technique to process the sensed data to determine a bounding polygon associated with the target object.

14. The system of claim 13 , wherein the sensor comprises a visual sensor.

15. The system of claim 1 , wherein the target object is being transported by a conveyor device.

16. A system, comprising:
a processor configured to:
obtain a plurality of trajectories corresponding to a plurality of detected objects being transported on a conveyor device, including to determine a trajectory associated with a detected object, wherein to determine the trajectory associated with the detected object comprises to:
obtain a plurality of image frames showing the detected object;
determine a set of velocities associated with a movement of the detected object based at least in part on the plurality of image frames and time intervals between image frames, wherein the set of velocities corresponds to a set of dimensions associated with a coordinate system; and
generate the trajectory associated with the detected object based at least in part on the set of velocities associated with the movement of the detected object;

obtain a global velocity of the conveyor device based at least in part on the plurality of trajectories; and
determine a new trajectory for a newly detected object based at least in part on the global velocity of the conveyor device; and

a memory coupled to the processor and configured to provide the processor with instructions.

17. The system of claim 16 , wherein the processor is further configured to generate a composite view of movements associated with the plurality of detected objects based at least in part on stitching together the plurality of trajectories corresponding to the plurality of detected objects.

18. The system of claim 16 , wherein the processor is further configured to determine a bounding polygon corresponding to the newly detected object, wherein a first portion of the bounding polygon is assigned a lower confidence value than a second portion of the bounding polygon.

19. The system of claim 18 , wherein the processor is further configured to modify a parameter of the bounding polygon based at least in part on an assigned confidence value associated with the parameter.

20. The system of claim 19 , where the assigned confidence value is determined at least in part by a machine learning algorithm.

21. The system of claim 16 , wherein the processor is further configured to determine a number of times that the newly detected object is observed by one or more sensors against a view threshold, and in the event that the number of times that the newly detected object is viewed is lower than the view threshold, discard a computed new trajectory for the newly detected object.

22. The system of claim 16 , wherein the plurality of trajectories is determined based at least in part on sensed data captured from a plurality of types of sensors with respect to the plurality of detected objects.

23. The system of claim 16 , wherein to determine the new trajectory for the newly detected object comprises to use the global velocity as an initial parameter of a dynamic movement model used to determine the new trajectory for the newly detected object.

24. A system, comprising:
a processor configured to:
select a first sorting device to sort a target object, wherein the target object is being transported along a conveyor device;
determine that the first sorting device has not successfully sorted the target object; and
in response to the determination that the first sorting device has not successfully sorted the target object, select a second sorting device to sort the target object, wherein to select the second sorting device comprises to:
determine a future location of the target object given a trajectory associated with the target object; and
select the second sorting device to sort the target object based at least in part on the future location of the target object and a location associated with the second sorting device; and

a memory coupled to the processor and configured to provide the processor with instructions.

25. The sy

Assignees

• Amp Robotics Corp

Inventors

• Horowitz Matanya B
• Castagneri Joseph M
• Schultz Carter J
• Taylor Kevin
• Anderson Austin
• Dawes Paul

Classifications

• G06V10/764

  using classification, e.g. of video objects · CPC title

• G06V10/62

  relating to a temporal dimension, e.g. time-based feature extraction; Pattern tracking · CPC title

• G06V20/50

  Context or environment of the image · CPC title

• G06T7/20

  Analysis of motion (motion estimation for coding, decoding, compressing or decompressing digital video signals H04N19/43, H04N19/51) · CPC title

• B65G2203/0208

  relating to the transported articles · CPC title