Self-driving vehicle collision management system

• US 10,259,452 B2
• Filed: 01/04/2017
• Issued: 04/16/2019
• Est. Priority Date: 01/04/2017
• Status: Active Grant

• Current Assignee: DoorDash, Inc.
• Original Assignee: International Business Machines Corporation
• Inventors: Gordon, Michael S., Hobson, Stacy F., Kozloski, James R., Kundu, Ashish, Pickover, Clifford A.
• Primary Examiner(s): Khatib, Rami
• Application Number:
  US15/397,855
• Publication Number:
  US 20180186366A1
• Time in Patent Office: 832 Days
• Field of Search: None
• CPC Class Codes: B60W 2530/10 Weight
• CPC Class Codes: B60W 2554/00 Input parameters relating to objects
• CPC Class Codes: B60W 30/085 Taking automatic action to adjust vehicle attitude in preparation for collision, e.g. braking for nose dropping
• CPC Class Codes: B60W 30/09 Taking automatic action to avoid collision, e.g. braking and steering

Abstract

A method generates and implements a real-time amelioration action for ameliorating an imminent collision between a self-driving vehicle (SDV) and an object. One or more processors detect that an imminent collision by a self-driving vehicle (SDV) is imminent with a confidence C1, and determine whether the SDV has an occupant of occupant type P with a confidence C2. One or more processors identify an object with which the imminent collision by the SDV is imminent with a confidence C3, and then generate and implement, based on C1, C2, C3, and P, a real-time amelioration action for ameliorating the imminent collision between the SDV and the object.

Claims

1. A computer-implemented method comprising:

detecting, by one or more processors, that an imminent collision by a self-driving vehicle (SDV) is imminent with a first confidence level, wherein the first confidence level is a first probability level that the SDV will hit an object if no ameliorative steps are taken by the SDV to avoid colliding with the object;

determining, by one or more processors, whether the SDV has an occupant of a particular occupant type with a second confidence level, wherein the particular occupant type is a single occupant type from a group of occupant types consisting of an animate occupant type and an inanimate occupant type, wherein the second confidence level is a second probability level that an occupant detection system on the SDV has accurately identified what type of occupant is currently within the SDV;

identifying, by one or more processors, an object with which the imminent collision by the SDV is imminent with a third confidence level, wherein the third confidence level is a third probability level that an object detection system on the SDV has accurately identified what type of object is about to be hit by the SDV if no ameliorative steps are taken by the SDV to avoid hitting the object; and

generating and implementing, by one or more processors and based on the first confidence level, the second confidence level, the third confidence level, and the particular occupant type, a real-time amelioration action for ameliorating the imminent collision between the SDV and the object, wherein the real-time amelioration action comprises adjusting a movement of the SDV.

2. The computer-implemented method of claim 1, further comprising:

determining, by one or more processors, the first confidence level based on an analysis of sensor data received in real-time from one or more sensors on the SDV.

3. The computer-implemented method of claim 1, wherein the object to be imminently collided with by the SDV is another SDV that has a human passenger.

4. The computer-implemented method of claim 1, wherein the object to be imminently collided with by the SDV is another SDV that has no passenger.

5. The computer-implemented method of claim 1, wherein the object to be imminently collided with by the SDV is a pedestrian.

6. The computer-implemented method of claim 1, wherein the object to be imminently collided with by the SDV is a vehicle that is not an SDV.

7. The computer-implemented method of claim 1, wherein the object to be imminently collided with by the SDV is an inanimate object located in a fixed position.

8. The computer-implemented method of claim 1, wherein the real-time amelioration action is to strike the object in a manner that causes energy-absorbing areas on the SDV to absorb an impact of the SDV striking the object.

9. The computer-implemented method of claim 1, wherein the SDV is unoccupied by an animate entity, and wherein the real-time amelioration action is to fill the interior of the SDV with a fast expanding foam to protect an interior of the SDV in response to the imminent collision between the SDV and the object.

10. The computer-implemented method of claim 1, wherein the SDV is unoccupied by an animate entity, and wherein the real-time amelioration action is to fill an interior of the SDV with a fire-retarding agent to inhibit an explosion and fire caused by the imminent collision between the SDV and the object.

11. The computer-implemented method of claim 1, wherein the SDV is a first SDV, wherein the object is a second SDV, and wherein the computer-implemented method further comprises:

coordinating, by one or more processors, movement between the first SDV and the second SDV in order to prevent the first SDV and the second SDV from hitting each other.

12. The computer-implemented method of claim 1, wherein the SDV is a first SDV, and wherein the computer-implemented method further comprises:

receiving, by one or processors, executable instructions for implementing an amelioration action performed by a group of other SDVs that experienced an imminent collision that was similar to the imminent collision being experienced by the first SDV; and

executing, by one or more processors in the first SDV, the executable instructions for implementing the amelioration action performed by the group of other SDVs.

13. The computer-implemented method of claim 1, wherein the method is implemented as a cloud-based service.

14. A computer program product for generating and implementing a real-time amelioration action for ameliorating an imminent collision between a self-driving vehicle (SDV) and an object, the computer program product comprising a non-transitory computer readable storage medium having program instructions embodied therewith, the program instructions readable and executable by a processor to cause the processor to:

detect that an imminent collision by a self-driving vehicle (SDV) is imminent with a first confidence level, wherein the first confidence level is a first probability level that the SDV will hit an object if no ameliorative steps are taken by the SDV to avoid colliding with the object;

determine whether the SDV has an occupant of a particular occupant type with a second confidence level, wherein the particular occupant type is a single occupant type from a group of occupant types consisting of an animate occupant type and an inanimate occupant type, wherein the second confidence level is a second probability level that an occupant detection system on the SDV has accurately identified what type of occupant is currently within the SDV;

identify an object with which the imminent collision by the SDV is imminent with a third confidence level, wherein the third confidence level is a third probability level that an object detection system on the SDV has accurately identified what type of object is about to be hit by the SDV if no ameliorative steps are taken by the SDV to avoid hitting the object; and

generate and implement, based on the first confidence level, the second confidence level, the third confidence level, and the particular occupant type, a real-time amelioration action for ameliorating the imminent collision between the SDV and the object, wherein the real-time amelioration action comprises adjusting a movement of the SDV.

15. The computer program product of claim 14, wherein the program instructions are provided as a service in a cloud environment.

16. A system comprising:

one or more processors;

one or more computer readable memories operably coupled to the one or more processors;

one or more computer readable storage mediums operably coupled to the one or more computer readable memories; and

program instructions stored on at least one of the one or more computer readable storage mediums for execution by at least one of the one or more processors via at least one of the one or more computer readable memories, the program instructions comprising;

program instructions configured to detect that an imminent collision by a self-driving vehicle (SDV) is imminent with a first confidence level, wherein the first confidence level is a first probability level that the SDV will hit an object if no ameliorative steps are taken by the SDV to avoid colliding with the object;

program instructions configured to determine whether the SDV has an occupant of a particular occupant type with a second confidence level, wherein the particular occupant type is from a group of occupant types consisting of animate occupant types and inanimate occupant types, wherein the second confidence level is a second probability level that an occupant detection system on the SDV has accurately identified what type of occupant is currently within the SDV;

program instructions configured to identify an object with which the imminent collision by the SDV is imminent with a third confidence level, wherein the third confidence level is a third probability level that an object detection system on the SDV has accurately identified what type of object is about to be hit by the SDV if no ameliorative steps are taken by the SDV to avoid hitting the object; and

program instructions configured to generate and implement, based on the first confidence level, the second confidence level, the third confidence level, and the particular occupant type, a real-time amelioration action for ameliorating the imminent collision between the SDV and the object, wherein the real-time amelioration action comprises adjusting a movement of the SDV.