Maintaining a data structure corresponding to a target object

Patent metadata

Field	Value
Publication number	US-12340287-B2
Application number	US-202117559818-A
Country	US
Kind code	B2
Filing date	Dec 22, 2021
Priority date	Dec 22, 2021
Publication date	Jun 24, 2025
Grant date	Jun 24, 2025

Abstract

Maintaining a data structure corresponding to a target object is disclosed, including: determining that an identified target object from a sensed signal is a new target object, wherein the sensed signal is generated at a sorting facility; generating a new data structure corresponding to the new target object; and updating the new data structure with an attribute associated with the new target object, wherein the attribute is derived from the sensed signal associated with the new target object.

First claim

What is claimed is:

1. A system, comprising:
an object recognition device configured to generate a sensed signal at a sorting facility;
a compute node comprising a processor configured to:
receive the sensed signal from the object recognition device;
determine that an identified target object from the sensed signal is a new target object;
generate a new data structure corresponding to the new target object;
update the new data structure with an attribute associated with the new target object, wherein the attribute is derived from the sensed signal associated with the new target object;
update the new data structure with a result of a first sorting operation that was performed on the new target object by a first sorting device; and
publish the new data structure to a message queue; and a second sorting device configured to:
read the new data structure from the message queue; and
perform a second sorting operation on the new target object based at least in part on the result of the first sorting operation and the attribute that are included in the new data structure,
wherein the second sorting device is configured to perform the second sorting operation on the new target object using one or more of the following sorting mechanisms: a robot with one or more actuatable arms, a picker mechanism, an array of air orifices, an array of paddles, and an array of pushing mechanisms.

2. The system of claim 1 , wherein to determine that the identified target object from the sensed signal is the new target object comprises to compare the attribute associated with the new target object with stored attributes stored in existing data structures corresponding to existing target objects.

3. The system of claim 2 , wherein the attribute comprises a bounding polygon associated with the new target object.

4. The system of claim 2 , wherein the attribute comprises a trajectory associated with the new target object.

5. The system of claim 4 , wherein the trajectory associated with the new target object is determined based at least in part on applying a dynamic model to the sensed signal associated with the new target object.

6. The system of claim 1 , wherein the sensed signal comprises a first sensed signal, wherein the attribute comprises a first attribute, and wherein the processor is further configured to update the new data structure with a second attribute associated with the new target object, wherein the second attribute is derived from a second sensed signal associated with the new target object.

7. The system of claim 6 , wherein the first attribute comprises a first current trajectory, and wherein the second attribute comprises a second current trajectory associated with the new target object, and wherein to update the new data structure with the second attribute associated with the new target object comprises to replace the first current trajectory with the second current trajectory in the new data structure.

8. The system of claim 6 , wherein the first attribute comprises a first bounding polygon associated with the new target object and is generated from the sensed signal, wherein the second attribute comprises a second bounding polygon associated with the new target object, wherein the second bounding polygon is generated by updating the first bounding polygon using boundary information derived from the second sensed signal.

9. The system of claim 6 , wherein the first sensed signal is output by a first sensor at a first time, where the second sensed signal is output by a second sensor at a second time, and wherein the second time is later than the first time.

10. The system of claim 6 , wherein the second sensed signal is associated with the second sorting operation that is performed on the new target object by the second sorting device, and wherein the second attribute describes whether the second sorting operation on the new target object was successful.

11. The system of claim 6 , wherein the second sensed signal is associated with a deposit operation that is performed on the new target object by the second sorting device, and wherein the second attribute describes whether the deposit operation with respect to the new target object has been completed.

12. The system of claim 11 , wherein the processor is further configured to:
determine that the second attribute describes that the deposit operation with respect to the new target object has been completed; and
in response to the determination that the second attribute describes that the deposit operation with respect to the new target object has been completed, mark the new data structure for reclamation.

13. The system of claim 1 , wherein the attribute comprises an object type associated with the new target object.

14. The system of claim 1 , wherein the attribute comprises a deposit location associated with the new target object.

15. The system of claim 1 , wherein the processor is configured to:
aggregate data associated with target objects based at least in part on data structures associated with the target objects; and
send the aggregated data associated with the target objects to a server associated with a manufacturer of at least a portion of the target objects.

16. The system of claim 1 , wherein to determine that the identified target object from the sensed signal is the new target object comprises to:
input data of the sensed signal into a machine learning model; and
receive a data structure relating to a type of target object.

17. The system of claim 1 , wherein the processor is configured to use a machine learning model with sensed data to identify that an object in the sorting facility is not a target object.

18. A method, comprising:
receiving a sensed signal from an object recognition device at a sorting facility;
determining that an identified target object from the sensed signal is a new target object;
generating a new data structure corresponding to the new target object;
updating the new data structure with an attribute associated with the new target object, wherein the attribute is derived from the sensed signal associated with the new target object;
updating the new data structure with a result of a first sorting operation that was performed on the new target object by a first sorting device;
publishing the new data structure to a message queue;
reading, via a second sorting device, the new data structure from the message queue; and
performing, via the second sorting device, a second sorting operation on the new target object based at least in part on the result of the first sorting operation and the attribute that are included in the new data structure,
wherein the second sorting device is configured to perform the second sorting operation on the new target object using one or more of the following sorting mechanisms: a robot with one or more actuatable arms, a picker mechanism, an array of air orifices, an array of paddles, and an array of pushing mechanisms.

19. The method of claim 18 , wherein determining that the identified target object from the sensed signal is the new target object comprises comparing the attribute associated with the new target object with stored attributes stored in existing data structures corresponding to existing target objects.

20. A computer program product, the computer program product being embodied in a non-transitory computer-readable storage medium and comprising computer instructions for:
receiving a sensed signal from an object recognition device at a sorting facility;
determining

Assignees

• Amp Robotics Corp

Inventors

• Horowitz Matanya B
• Castagneri Joseph M
• Browning Joshua M
• Anderson Austin
• Potter Carson C
• Dawes Paul

Classifications

• G06F16/5854

  using shape and object relationship · CPC title

• G06N20/00 Primary

  Machine learning · CPC title