Method and system for impact-based operation of an autonomous agent

Patent metadata

Field	Value
Publication number	US-12441364-B2
Application number	US-202418654502-A
Country	US
Kind code	B2
Filing date	May 3, 2024
Priority date	Dec 1, 2021
Publication date	Oct 14, 2025
Grant date	Oct 14, 2025

Abstract

A system for impact-based operation of an autonomous agent (equivalently referred to herein as an ego agent and autonomous vehicle) includes and/or interfaces with a computing subsystem (equivalently referred to herein as a computer and/or set of computers). A method for impact-based operation of an autonomous agent includes: receiving a set of inputs; predicting a set of future scenarios; and determining a set of metrics based on the set of future scenarios. Additionally or alternatively, the method can include operating the autonomous agent based on the set of metrics and/or any other processes.

First claim

We claim:

1. A method for operation of a vehicle, the method comprising:
detecting a set of objects in an environment of the vehicle, wherein the set of objects are separate and distinct from the vehicle;
using a multi-policy decision-making simulator, performing a set of simulations representing the set of objects to produce a set of metrics, the set of simulations comprising:
a 1 st subset of multiple simulations, wherein each of the 1 st subset of simulations represents the set of objects and the vehicle; and
a 2 nd subset of multiple simulations, wherein the 2 nd subset of simulations represents only the set of objects, wherein each of the 2 nd subset of simulations comprises a set of intent assignments for each of the set of objects, wherein the set of intent assignments for each respective object differs between simulations of the 2 nd subset of multiple simulations;

determining an impact score of the vehicle on the set of objects in the environment based on the set of metrics; and
operating the vehicle based on the impact score.

2. The method of claim 1 , wherein the set of metrics is determined based on:
a 1 st subset of metrics calculated based on outcomes of the 1 st subset of multiple simulations; and
a 2 nd subset of metrics calculated based on outcomes of the 2 nd subset of multiple simulations.

3. The method of claim 2 , wherein producing the set of metrics comprises comparing the 1 st subset of metrics with the 2 nd subset of metrics.

4. The method of claim 3 , wherein comparing the 1 st subset of metrics with the 2 nd subset of metrics comprises calculating a difference between the 1 st and 2 nd subset of metrics.

5. The method of claim 1 , wherein determining the impact score comprises calculating the impact score based on the set of metrics.

6. The method of claim 1 , wherein:
a representation of the vehicle is present in the 1 st subset of multiple simulations; and
the representation of the vehicle is absent in the 2 nd subset of multiple simulations.

7. The method of claim 1 , wherein each of the set of simulations is a forward simulation.

8. The method of claim 7 , wherein each of the set of simulations simulates a predetermined time period in the future, wherein a duration of the time period is equal among all simulations of the set.

9. The method of claim 1 , wherein the set of simulations is associated with a set of potential policies of the vehicle, wherein each potential policy of the vehicle is associated with a single simulation of the 1 st subset of multiple simulations and a single simulation of the 2 nd subset of multiple simulations, wherein operating the vehicle based on the impact score comprises selecting a single potential policy from the set based on the impact score.

10. The method of claim 9 , wherein selecting a first policy from the set of potential policies of the vehicle based on the impact score comprises determining a policy-specific impact score for each potential policy of the set of potential policies of the vehicle and selecting the first policy from the set of potential policies of the vehicle based on a respective policy-specific impact score.

11. A system for operation of a vehicle, the system comprising:
a processing subsystem configured to detect a set of objects in an environment of the vehicle, wherein the set of objects are separate and distinct from the vehicle;
a multi-policy decision-making simulator that performs a set of simulations, the set of simulations comprising:
a 1 st subset of multiple simulations wherein the 1 st subset of multiple simulations represents the set of objects and the vehicle; and
a 2 nd subset of multiple simulations, wherein the 2 nd subset of multiple simulations represents only the set of objects, wherein each of the 2 nd subset of multiple simulations comprises a set of intent assignments for each of the set of objects, wherein the set of intent assignments for each respective object differs between simulations of the 2 nd subset of multiple simulations;

a control subsystem configured to maneuver the vehicle based on instructions from the processing subsystem, wherein the processing subsystem:
produces a set of metrics based on the set of simulations;
based on the set of metrics, determines an impact score corresponding to an impact of the vehicle on the set of objects in the environment; and
produces the instructions based at least in part on the potential impact score.

12. The system of claim 11 , wherein the set of metrics is determined based on:
a 1 st subset of metrics calculated based on outcomes of the 1 st subset of multiple simulations; and
a 2 nd subset of metrics calculated based on outcomes of the 2 nd subset of multiple simulations.

13. The system of claim 12 , wherein the processing subsystem is configured to produce the set of metrics based on comparing the 1 st subset of metrics with the 2 nd subset of metrics.

14. The system of claim 13 , wherein comparing the 1 st subset of metrics with the 2 nd subset of metrics comprises calculating a difference between the 1 st and 2 nd subset of metrics.

15. The system of claim 11 , wherein:
a representation of the vehicle is present in the 1 st subset of multiple simulations; and
the representation of the vehicle is absent in the 2 nd subset of multiple simulations.

16. The system of claim 11 , wherein each of the set of simulations is a forward simulation.

17. The system of claim 16 , wherein each of the set of simulations simulates a predetermined time period in the future, wherein a duration of the time period is equal among all simulations of the set.

18. The system of claim 11 , wherein the set of simulations is associated with a set of potential policies, wherein each potential policy of the set is associated with a single simulation of the 1 st subset of multiple simulations and a single simulation of the 2 nd subset of multiple simulations, wherein maneuvering the vehicle based on the instructions comprises selecting a single potential policy of the set based on the impact score.

19. The system of claim 18 , wherein determining the impact score comprises determining a respective impact score for each potential policy of the set of potential policies.

Assignees

• May Mobility Inc

Inventors

• Johnson Collin
• Bagla Yash
• Borle Kapil
• Katpatal Akshay
• Mahmood Mahmood
• Olson Edwin B

Classifications

• G05D1/81

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

• B60W60/0055

  only part of driving tasks shifted to occupants · CPC title

• B60W60/005

  Handover processes (Handing over between remote control and on-board control or handing over between remote control arrangements G05D1/227) · CPC title

• B60W50/0097

  Predicting future conditions · CPC title

• B60W2050/0028

  Mathematical models, e.g. for simulation · CPC title