Predicting behaviors of road agents using intermediate intention signals

Patent metadata

Field	Value
Publication number	US-12168462-B2
Application number	US-202318228365-A
Country	US
Kind code	B2
Filing date	Jul 31, 2023
Priority date	Nov 23, 2020
Publication date	Dec 17, 2024
Grant date	Dec 17, 2024

Abstract

An autonomous vehicle includes sensor subsystem(s) that output a sensor signal. A perception subsystem (i) detects an agent in a vicinity of the autonomous vehicle and (ii) generates a motion signal that describes at least one of a past motion or a present motion of the agent. An intention prediction subsystem processes the sensor signal to generate an intention signal that describes at least one intended action of the agent. A behavior prediction subsystem processes the motion signal and the intention signal to generate a behavior prediction signal that describes at least one predicted behavior of the agent. A planner subsystem processes the behavior prediction signal to plan a driving decision for the autonomous vehicle.

First claim

What is claimed is:

1. A computer-implemented method, comprising:
identifying an intended action of a road agent detected by an autonomous vehicle, the intended action representing an action that the road agent would take if other agents detected in a vicinity of the road agent were disregarded;
identifying a past movement of the road agent;
predicting a behavior of the road agent based on the intended action of the road agent and the past movement of the road agent;
generating a driving decision for the autonomous vehicle based on the predicted behavior of the road agent; and
initiating a movement of the autonomous vehicle based on the driving decision.

2. The method of claim 1 , wherein the road agent is a pedestrian or a cyclist, and the intended action describes an intention of the pedestrian or the cyclist to at least one of:
(i) cross a roadway on which the autonomous vehicle is traveling,
(ii) not cross the roadway, or
(iii) roam within a defined area in proximity to the roadway.

3. The method of claim 1 , wherein the road agent is a vehicle other than the autonomous vehicle, and the intended action describes an intention of the vehicle to at least one of:
(i) move from one lane of a roadway on which the vehicle is traveling to another lane of the roadway,
(ii) yield to another agent on the roadway,
(iii) come to a stop on the roadway,
(iv) accelerate,
(v) decelerate, or
(vi) make a turn onto another roadway.

4. The method of claim 1 , wherein the intended action further represents a prediction of action that the road agent would take if the road agent disregarded a presence of the autonomous vehicle in the environment.

5. The method of claim 1 , wherein the predicted behavior of the road agent is predicted based on a behavior of the autonomous vehicle.

6. The method of claim 1 , wherein identifying the past movement of the road agent comprises identifying at least one of a past heading of the road agent at one or more past points in time, a past speed of the road agent at the one or more past points in time, or a past acceleration of the road agent at the one or more past points in time.

7. The method of claim 1 , wherein the intended action of the road agent is determined by an intention prediction subsystem that is more computationally demanding than a a behavior prediction subsystem that predicts the behavior of the road agent.

8. A system, comprising:
one or more processors; and
one or more computer-readable media having instructions stored thereon that, when executed by the one or more processors, cause performance of operations comprising:
identifying an intended action of a road agent detected by an autonomous vehicle, the intended action representing an action that the road agent would take if other agents detected in a vicinity of the road agent were disregarded;
identifying a past movement of the road agent;
predicting a behavior of the road agent based on the intended action of the road agent and the past movement of the road agent;
generating a driving decision for the autonomous vehicle based on the predicted behavior of the road agent; and
initiating a movement of the autonomous vehicle based on the driving decision.

9. The system of claim 8 , wherein the road agent is a pedestrian or a cyclist, and the intended action describes an intention of the pedestrian or the cyclist to at least one of:
(i) cross a roadway on which the autonomous vehicle is traveling,
(ii) not cross the roadway, or
(iii) roam within a defined area in proximity to the roadway.

10. The system of claim 8 , wherein the road agent is a vehicle other than the autonomous vehicle, and the intended action describes an intention of the vehicle to at least one of:
(i) move from one lane of a roadway on which the vehicle is traveling to another lane of the roadway,
(ii) yield to another agent on the roadway,
(iii) come to a stop on the roadway,
(iv) accelerate,
(v) decelerate, or
(vi) make a turn onto another roadway.

11. The system of claim 8 , wherein the intended action further represents a prediction of action that the road agent would take if the road agent disregarded a presence of the autonomous vehicle in the environment.

12. The system of claim 8 , wherein the predicted behavior of the road agent is predicted based on a behavior of the autonomous vehicle.

13. The system of claim 8 , wherein identifying the past movement of the road agent comprises identifying at least one of a past heading of the road agent at one or more past points in time, a past speed of the road agent at the one or more past points in time, or a past acceleration of the road agent at the one or more past points in time.

14. The system of claim 8 , wherein the intended action of the road agent is determined by an intention prediction subsystem that is more computationally demanding than a behavior prediction subsystem that predicts the behavior of the road agent.

15. One or more non-transitory computer-readable media having instructions stored thereon that, when executed by one or more processors, cause performance of operations comprising:
identifying an intended action of a road agent detected by an autonomous vehicle, the intended action representing an action that the road agent would take if other agents detected in a vicinity of the road agent were disregarded;
identifying a past movement of the road agent;
predicting a behavior of the road agent based on the intended action of the road agent and the past movement of the road agent;
generating a driving decision for the autonomous vehicle based on the predicted behavior of the road agent; and
initiating a movement of the autonomous vehicle based on the driving decision.

16. The one or more non-transitory computer-readable media of claim 15 , wherein the road agent is a pedestrian or a cyclist, and the intended action describes an intention of the pedestrian or the cyclist to at least one of:
(i) cross a roadway on which the autonomous vehicle is traveling,
(ii) not cross the roadway, or
(iii) roam within a defined area in proximity to the roadway.

17. The one or more non-transitory computer-readable media of claim 15 , wherein the road agent is a vehicle other than the autonomous vehicle, and the describes an intention of the vehicle to at least one of:
(i) move from one lane of a roadway on which the automotive vehicle is traveling to another lane of the roadway,
(ii) yield to another agent on the roadway,
(iii) come to a stop on the roadway,
(iv) accelerate,
(v) decelerate, or
(vi) make a turn onto another roadway.

18. The one or more non-transitory computer-readable media of claim 15 , wherein the intended action further represents a prediction of action that the road agent would take if the road agent disregarded a presence of the autonomous vehicle in the environment.

19. The one or more non-transitory computer-readable media of claim 15 , wherein the predicted behavior of the road agent is predicted based on a behavior of the autonomous vehicle.

20. The one or more non-transitory computer-readable media of claim 15 , wherein identifying the past movement of the road agent comprises identifying at least one of a past heading of the road agent at one or more past points in time, a past speed of the road agent at the one or more past points in time, or a past acceleration of the road agent at the one or more past points in time.

Assignees

• Waymo Llc

Inventors

• Refaat Khaled

Classifications

• B60W2420/408

  Radar; Laser, e.g. lidar · CPC title

• G01S13/931

  of land vehicles · CPC title

• G01S13/58

  Velocity or trajectory determination systems; Sense-of-movement determination systems · CPC title

• B60W2555/00

  Input parameters relating to exterior conditions, not covered by groups B60W2552/00, B60W2554/00 · CPC title

• G01S17/58

  Velocity or trajectory determination systems; Sense-of-movement determination systems · CPC title