Multi-perspective system and method for behavioral policy selection by an autonomous agent

Patent metadata

Field	Value
Publication number	US-11269332-B2
Application number	US-202117181302-A
Country	US
Kind code	B2
Filing date	Feb 22, 2021
Priority date	Jul 20, 2018
Publication date	Mar 8, 2022
Grant date	Mar 8, 2022

Abstract

A system and a method for autonomous decisioning and operation by an autonomous agent includes: collecting decisioning data including: collecting a first stream of data includes observation data obtained by onboard sensors of the autonomous agent, wherein each of the onboard sensors is physically arranged on the autonomous agent; collecting a second stream of data includes observation data obtained by offboard infrastructure devices, the offboard infrastructure devices being arranged geographically remote from and in an operating environment of the autonomous agent; implementing a decisioning data buffer that includes the first stream of data from the onboard sensors and the second stream of data from the offboard sensors; generating current state data; generating/estimating intent data for each of one or more agents within the operating environment of the autonomous agent; identifying a plurality of candidate behavioral policies; and selecting and executing at least one of the plurality of candidate behavioral policies.

First claim

What is claimed is:

1. A system for operating an autonomous agent, the system comprising:
a plurality of offboard infrastructure devices configured to collect a first set of observation data;
a communication interface of the autonomous agent in communication with the plurality of offboard infrastructure devices;
an onboard sensor system arranged onboard the autonomous agent and configured to collect a second set of observation data;
a computing system comprising:
an onboard computing subsystem arranged onboard the autonomous agent; and
a distributed network of offboard computing devices in communication with the set of offboard infrastructure devices;

wherein the computing system is configured to, for each of a set of multiple time points during operation of the autonomous agent:
identify a set of multiple candidate behavioral policies for the autonomous agent based on the first and second sets of observation data;
select a behavioral policy from the set of multiple candidate behavioral policies based on a set of criteria; and
control the autonomous agent according to the behavioral policy.

2. The system of claim 1 , wherein the computing system is further configured to perform, based on the first and second sets of observation data:
tracking one or more agents in an environment of the autonomous agent; and
intent estimation of the one or more agents;
wherein the behavioral policy is selected further based on the tracking and the intent estimation.

3. The system of claim 2 , wherein the distributed network of offboard computing devices is configured to perform at least a portion of at least one of the tracking and the intent estimation.

4. The system of claim 3 , wherein the distributed network of offboard computing devices is configured to perform at least a portion of each of the tracking and the intent estimation.

5. The system of claim 1 , wherein each of the plurality of offboard infrastructure devices is associated with a field-of-sensing region proximal to the offboard infrastructure device.

6. The system of claim 5 , wherein a geometry of the field-of-sensing region is predetermined.

7. The system of claim 1 , wherein the distributed network of offboard computing devices is arranged onboard the set of offboard infrastructure devices.

8. The system of claim 1 , wherein the distributed network of offboard computing devices is arranged remote from the autonomous agent.

9. A method for operating an autonomous agent, the method comprising:
collecting a first set of observation data at an autonomous agent;
detecting that the autonomous agent is located within a communication proximity associated with a set of offboard infrastructure devices;
in response to detecting that the autonomous agent is located within the communication proximity, collecting a second set of observation data from the set of offboard infrastructure devices, wherein the set of offboard infrastructure devices is arranged remotely from the autonomous agent, and wherein each of the set of offboard infrastructure devices is associated with a field of sensing;
simulating each of a plurality of candidate behavioral policies for the autonomous agent based on the first and second sets of observation data;
selecting a behavioral policy from the plurality of candidate behavioral policies based on simulating each of the plurality of candidate behavioral policies and a set of criteria; and
controlling the autonomous agent according to the behavioral policy.

10. The method of claim 9 , wherein the first set of observation data is collected from a sensor system onboard the autonomous agent.

11. The method of claim 9 , wherein the field of sensing associated with each of the set of offboard infrastructure devices is predetermined.

12. The method of claim 11 , wherein the field of sensing associated with each of the set of offboard infrastructure devices is a geometrically-defined region.

13. The method of claim 9 , wherein collecting the second set of observation data comprises collecting abstractions of objects located within the fields of sensing of the set of offboard infrastructure devices.

14. A method for operating an autonomous agent, the method comprising:
collecting a first set of observation data at an autonomous agent;
collecting a second set of observation data at a set of offboard infrastructure devices, wherein the set of offboard infrastructure devices is arranged remotely from the autonomous agent, and wherein each of the set of offboard infrastructure devices is associated with a field of sensing;
for each of a set of multiple time points during operation of the autonomous agent:
identifying a plurality of candidate behavioral policies for the autonomous agent based on the first and second sets of observation data;
selecting a behavioral policy from the plurality of candidate behavioral policies based on a set of criteria; and
controlling the autonomous agent according to the behavioral policy.

15. The method of claim 14 , wherein the first set of observation data is collected from a sensor system onboard the autonomous agent.

16. The method of claim 14 , wherein the field of sensing associated with each of the set of offboard infrastructure devices is predetermined.

17. The method of claim 16 , wherein the field of sensing associated with each of the set of offboard infrastructure devices is a geometrically-defined region.

18. The method of claim 14 , wherein collecting the second set of observation data comprises collecting abstractions of objects located within the fields of sensing of the set of offboard infrastructure devices.

19. The method of claim 18 , wherein the second set of observation data is automatically collected in response to detecting that the autonomous agent is located within a communication proximity associated with at least one of the set of offboard infrastructure devices.

20. The method of claim 18 , wherein the fields of sensing of the set of offboard infrastructure devices are at least partially overlapping with a field of sensing of a sensor system onboard the autonomous agent.

Assignees

• May Mobility Inc

Inventors

• Vozar Steve
• Olson Edwin
• Voorheis Tom

Classifications

• G05D1/81

  Handing over between on-board automatic and on-board manual control · CPC title

• G07C5/008 Primary

  communicating information to a remotely located station (transmission systems for measured values G08C) · CPC title

• G08G1/0112

  from the vehicle, e.g. floating car data [FCD] · CPC title

• B60W2554/4029

  Pedestrians · CPC title

• B60W60/0027

  using trajectory prediction for other traffic participants · CPC title