Autonomous vehicle maneuvering based upon risk associated with occluded regions

Patent metadata

Field	Value
Publication number	US-11767011-B2
Application number	US-202217850994-A
Country	US
Kind code	B2
Filing date	Jun 27, 2022
Priority date	Mar 31, 2019
Publication date	Sep 26, 2023
Grant date	Sep 26, 2023

Abstract

An autonomous vehicle (AV) is described herein. The AV is configured to identify an occluded region where a portion of a field of view of a sensor is occluded by an object. The AV is further configured to hypothesize that an object exists in the occluded region and is moving in the occluded region. The AV is still further configured to perform a driving maneuver based upon the hypothesized object existing in the occluded region.

First claim

What is claimed is:

1. An autonomous vehicle (AV) comprising:
a computing system;
a sensor system that is in communication with the computing system; and
a mechanical system that is operably coupled to the computing system, wherein the computing system is programmed to perform acts comprising:
defining an occluded region based upon a sensor signal output by the sensor system, wherein the occluded region is a spatial region in an environment of the AV where the AV lacks visibility;
at a first point in time, inferring that an object is within the occluded region and is travelling towards the AV;
inferring that, at a second point in time that is subsequent the first point in time, the object will begin to decelerate as a reaction to the AV becoming visible to the object; and
controlling the mechanical system of the AV based upon the inferring that the object will begin to decelerate at the second point in time based on the reaction to the AV becoming visible to the object.

2. The AV of claim 1 , the acts further comprising:
inferring, at the first point in time, a velocity of the object that is inferred to be travelling towards the AV.

3. The AV of claim 2 , the acts further comprising:
inferring that the object will travel at the velocity towards the AV until the second point in time.

4. The AV of claim 1 , the acts further comprising:
inferring a deceleration rate of the object.

5. The AV of claim 4 , wherein the object is inferred to decelerate at a constant rate subsequent to the second point in time.

6. The AV of claim 1 , the acts further comprising computing a position in the environment of the AV where the object will come to a stop, wherein the position is computed based upon the inferring that the object will begin to decelerate at the second point in time.

7. The AV of claim 6 , wherein the position is computed based further upon a difference between the first point in time and the second point in time.

8. The AV of claim 1 , wherein controlling the mechanical system of the AV comprises causing the AV to enter a lane of a roadway within which traffic flows in a direction that is opposite a direction of travel of the AV.

9. The AV of claim 1 , wherein controlling the mechanical system of the AV comprises causing the AV to turn across a lane of traffic within which traffic flows in a direction that is opposite a direction of travel of the AV.

10. The AV of claim 1 , wherein controlling the mechanical system of the AV is further based upon a velocity of the AV at the first point in time.

11. A method performed by an autonomous vehicle (AV), the method comprising:
generating, by a sensor system, a sensor signal, wherein the sensor signal is indicative of objects in an environment surrounding the AV;
based upon the sensor signal, defining an occluded region, wherein the occluded region is a spatial region in the environment surrounding the AV within which the AV lacks visibility;
upon defining the occluded region, and at a current point in time, inferring that an object is at a position within the occluded region and is travelling towards the AV;
inferring that the object will begin to decelerate at a future point in time and prior to reaching the AV as a reaction to the AV becoming visible to the object; and
performing a driving maneuver based upon the inferring that the object will begin to decelerate at the future point in time based on the reaction to the AV becoming visible to the object.

12. The method of claim 11 , further comprising:
computing a position where the object will come to a stop based upon the inferred position of the object within the occluded region and a rate of deceleration assigned to the object, wherein the driving maneuver is performed based upon the computed position where the object will come to a stop.

13. The method of claim 12 , further comprising assigning a velocity to the object at the position of the object within the occluded region, wherein the position where the object will come to a stop is computed based further upon the velocity assigned to the object.

14. The method of claim 13 , further comprising inferring that the object will travel towards the AV at the velocity assigned to the object from the current time to the future point in time, wherein the position where the object will come to a stop is computed based further upon the inferring that the object will travel towards the AV at the velocity assigned to the object from the current time to the future point in time.

15. The method of claim 11 , wherein the object has a deceleration rate assigned thereto, the method further comprising inferring that the object will decelerate at the deceleration rate until the object comes to a stop.

16. The method of claim 11 , wherein the sensor system is a LIDAR sensor system.

17. The method of claim 11 , wherein the driving maneuver is a turn across a lane of traffic.

18. The method of claim 11 , wherein the driving maneuver is entry into a lane of oncoming traffic.

19. An autonomous vehicle (AV) comprising a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium comprises instructions that, when executed by a processor, cause the processor to perform acts comprising:
receiving a sensor signal generated by a sensor system of the AV;
based upon the sensor signal, defining an occluded region, wherein the occluded region is a spatial region in an environment surrounding the AV within which the AV lacks visibility;
upon defining the occluded region, and at a current point in time, inferring that an object is at a position within the occluded region and is travelling towards the AV;
inferring that the object will begin to decelerate at a future point in time and prior to reaching the AV as a reaction to the AV becoming visible to the object; and
controlling a mechanical system of the AV to cause the AV to perform a driving maneuver, wherein the AV is controlled based upon the inferring that the object will begin to decelerate at the future point in time based on the reaction to the AV becoming visible to the object.

20. The AV of claim 19 , the acts further comprising inferring that the object will continue to decelerate at a predefined deceleration rate subsequent to the future point in time, wherein the mechanical system of the AV is controlled based further upon the inferring that the object will continue to decelerate at the predefined deceleration rate.

Assignees

• Gm Cruise Holdings Llc

Inventors

• Ghose Shiva
• Chen Daniel
• Nitz Ted
• Philippona Christophe
• Harrington Curt
• Daily-Diamond Christopher
• Swofford Mason
• Holben Ryan
• Lujan Eric
• Greer Benjamin
• Capodieci Francesco

Classifications

• G05D1/617

  Safety or protection, e.g. defining protection zones around obstacles or avoiding hazards (arrangements for controlling the position or course of two or more vehicles for avoiding collisions therebetween G05D1/693; arrangements for reacting to or preventing system or operator failure G05D1/80) · CPC title

• G05D1/81

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

• B60W30/09 Primary

  Taking automatic action to avoid collision, e.g. braking and steering · CPC title

• B60W30/0956

  the prediction being responsive to traffic or environmental parameters · CPC title

• G05D1/0088

  characterized by the autonomous decision making process, e.g. artificial intelligence, predefined behaviours (using knowledge based models G06N5/00) · CPC title