Method and system for using a reaction of other road users to ego-vehicle actions in autonomous driving

Patent metadata

Field	Value
Publication number	US-11648965-B2
Application number	US-202017034366-A
Country	US
Kind code	B2
Filing date	Sep 28, 2020
Priority date	Sep 28, 2020
Publication date	May 16, 2023
Grant date	May 16, 2023

Abstract

A method of determining a trajectory for an autonomous vehicle is disclosed. An ego-vehicle may detect a moving actor in an environment. To choose between candidate trajectories for the ego-vehicle, the system will consider the cost of each candidate trajectory to the moving actor. The system will use the candidate trajectory costs for the candidate trajectories to select one of the candidate trajectories via which to move the ego-vehicle. An autonomous vehicle system of the ego-vehicle may then move the ego-vehicle in the environment along the selected trajectory.

First claim

The invention claimed is:

1. A method of determining a trajectory for an autonomous vehicle, the method comprising:
by a presence detection system of an ego-vehicle, detecting presence of a moving actor in an environment while the ego-vehicle is moving about the environment;
identifying a plurality of candidate trajectories for the ego-vehicle;
for each of the candidate trajectories of the ego-vehicle:
determining a nominal trajectory prediction that comprises a nominal trajectory that the moving actor would be expected to follow in reaction to past actions of the ego-vehicle,
determining a reactive trajectory that the moving actor is expected to follow in reaction to the candidate trajectory of the ego-vehicle,
determining a cost to the moving actor for the reactive trajectory, wherein the cost to the moving actor is a function of a lateral offset of the reactive trajectory from the nominal trajectory and a distance along a curve of a deviation of the reactive trajectory from the nominal trajectory, and
determining a candidate trajectory cost for the candidate trajectory, wherein the candidate trajectory cost is a function of a cost to the moving actor for the reactive trajectory and one or more ego-centric factors;

using the candidate trajectory costs for the candidate trajectories to select one of the candidate trajectories via which to move the ego-vehicle; and
by an autonomous vehicle system of the ego-vehicle, moving the ego-vehicle in the environment along the selected trajectory.

2. The method of claim 1 , wherein using the candidate trajectory costs for the candidate trajectories to select one of the candidate trajectories comprises selecting the candidate trajectory having the lowest candidate trajectory cost.

3. The method of claim 1 , wherein using the candidate trajectory costs for the candidate trajectories to select one of the candidate trajectories comprises selecting, from the candidate trajectories, a trajectory having a lowest cost to the moving actor.

4. The method of claim 1 , wherein the cost to the moving actor is a function of either or both of the following:
severity of a control action associated with the reactive trajectory; or
how close the reactive trajectory will come to the candidate trajectory.

5. The method of claim 1 , wherein determining the reactive trajectory that the moving actor is expected to follow in reaction to the candidate trajectory of the ego-vehicle comprises:
determining a plurality of possible reactive trajectories that the moving actor may follow in reaction to the candidate trajectory;
determining a reactive trajectory cost for each of the possible reactive trajectories; and
selecting, as the reactive trajectory that the moving actor is expected to follow, the possible reactive trajectory having the lowest reactive trajectory cost.

6. The method of claim 1 , wherein determining the candidate trajectory costs for the candidate trajectories is performed by a motion planning system of the ego-vehicle.

7. The method of claim 1 , wherein determining the candidate trajectory costs for the candidate trajectories is performed by an external system that is in communication with the ego-vehicle.

8. A system for determining a trajectory for autonomous vehicle, the system comprising:
a presence detection system that is capable of detecting presence of a moving actor in an environment while the ego-vehicle is moving about the environment;
a processor; and
a memory containing programming instructions that are configured to cause the processor to:
identify a plurality of candidate trajectories for the ego-vehicle,
for each of the candidate trajectories of the ego-vehicle:
determine a nominal trajectory prediction that comprises a nominal trajectory that the moving actor would be expected to follow in reaction to past actions of the ego-vehicle;
determine a reactive trajectory that the moving actor is expected to follow in reaction to the candidate trajectory of the ego-vehicle;
determine a cost to the moving actor for the reactive trajectory, wherein the cost to the moving actor is a function of a lateral offset of the reactive trajectory from the nominal trajectory and a distance along a curve of a deviation of the reactive trajectory from the nominal trajectory; and
determine a candidate trajectory cost for the candidate trajectory, wherein the candidate trajectory cost is a function of a cost to the moving actor for the reactive trajectory and one or more ego-centric factors, and

use the candidate trajectory costs for the candidate trajectories to select one of the candidate trajectories via which to move the ego-vehicle, and
cause an autonomous vehicle system of the ego-vehicle to move the ego-vehicle in the environment along the selected trajectory.

9. The system of claim 8 , further comprising the autonomous vehicle system of the ego-vehicle, wherein the autonomous vehicle system comprises an onboard computing device and programming that is configured to cause the ego-vehicle to move along the selected trajectory.

10. The system of claim 8 , wherein the presence detection system comprises one or more of the following: a camera or a LiDAR system.

11. The system of claim 8 , wherein the instructions to use the candidate trajectory costs for the candidate trajectories to select one of the candidate trajectories comprise instructions to select the candidate trajectory having the lowest candidate trajectory cost.

12. The system of claim 8 , wherein the instructions to use the candidate trajectory costs for the candidate trajectories to select one of the candidate trajectories comprise instructions to select, from the candidate trajectories, a trajectory having a lowest cost to the moving actor.

13. The system of claim 8 , wherein the cost to the moving actor is a function of either or both of the following:
severity of a control action associated with the reactive trajectory; or
how close the reactive trajectory will come to the candidate trajectory.

14. The system of claim 8 , wherein the instructions to determine the reactive trajectory that the moving actor is expected to follow in reaction to the candidate trajectory of the ego-vehicle comprise instructions to:
determine a plurality of possible reactive trajectories that the moving actor may follow in reaction to the candidate trajectory;
determine a reactive trajectory cost for each of the possible reactive trajectories; and
select, as the reactive trajectory that the moving actor is expected to follow, the possible reactive trajectory having the lowest reactive trajectory cost.

15. The system of claim 8 , wherein the processor is a component of a motion planning system of the ego-vehicle.

16. The system of claim 8 , wherein the processor is a component of an external system that is in communication with the ego-vehicle.

17. A method of determining a trajectory for an autonomous vehicle, the method comprising:
by a presence detection system of an ego-vehicle, detecting presence of a moving actor in an environment while the ego-vehicle is moving about the environment;
identifying a plurality of candidate trajectories for the ego-vehicle;
for each of the candidate trajectories of the ego-vehicle:
determining a reactive trajectory that the moving actor is expected to follow in reaction to the candidate trajectory of the ego-vehicle,
determining a nominal trajectory prediction that comprises a nominal trajectory that the moving actor would be expected to follow in reaction to past actions of the ego-vehicle,
determining a cost to the moving

Assignees

• Argo Ai Llc

Inventors

• Cunningham Christopher
• Seegmiller Neal
• Ollis Mark
• Hartnett Andrew T
• Carr G Peter K
• Savtchenko Constantin

Classifications

• G01C21/3453 Primary

  Special cost functions, i.e. other than distance or default speed limit of road segments · CPC title

• B60W60/0027 Primary

  using trajectory prediction for other traffic participants · CPC title

• B60W2556/55

  using telemetry · CPC title

• B60W2554/802

  Longitudinal distance · CPC title

• B60W60/0011

  involving control alternatives for a single driving scenario, e.g. planning several paths to avoid obstacles · CPC title