Imminent-collision detection system and process

• US 7,409,295 B2
• Filed: 08/09/2004
• Issued: 08/05/2008
• Est. Priority Date: 08/09/2004
• Status: Active Grant

• Current Assignee: Arriver Software LLC
• Original Assignee: M/A Com
• Inventors: Paradie, Michael John
• Primary Examiner(s): Hughes; Deandra M.
• Assistant Examiner(s): Diacou; Ari M
• Application Number:
  US10/914,360
• Publication Number:
  US 20060031015A1
• Time in Patent Office: 1,457 Days
• Field of Search: 701/301, 701 45- 47
• US Class Current: 701/301
• CPC Class Codes: B60R 21/0134 responsive to imminent contact with an obstacle , e.g. using radar systems
• CPC Class Codes: G01S 5/16 using electromagnetic waves other than radio waves
• CPC Class Codes: G06T 7/70 Determining position or orientation of objects or cameras camera calibration G06T7/80
• CPC Class Codes: G06V 20/58 Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads

Abstract

A process of determining an imminent-collision between a vehicle and an object, said vehicle having a sensing system for obtaining one or more images representing at least one observed object within a field of detection, said process comprising: (a) obtaining one or more images representing an observed object within said field of detection; and (b) determining that a collision between said vehicle and said observed object is imminent when the ratio of the probability that said observed object is actually within a collision zone to the probability that said observed object is actually within a safe zone is greater than a certain value.

Claims

1. A process of determining an imminent-collision between a vehicle and an object, said vehicle having a sensing system for obtaining position data on at least one observed object within a field of detection, said process comprising:

(a) obtaining position data on an object within said field of detection;

(b) determining that said object is likely to be between first and second collision courses when the following two conditions are met, (1) a probability density function value that said object is on said first collision course exceeds a probability density function value of said object being on a first avoidance course, and (2) a probability density function value that said object is on said second collision course exceeds a probability density function value of said object being on a second avoidance course,

wherein said first avoidance course is on one side of said first and second collision courses, and said second avoidance course is on the other side of said first and second collision courses; and

(c) providing information to facilitate deployment of collision mitigation measures if said object is determined to be between said first and second collision courses.

2. The process of claim 1, wherein, step (b) comprises determining a first ratio of said probability density function value that said object is on said first collision course to said probability density value that said object is on said first avoidance course and a second ratio of said probability density function value that said object is on said second collision course to said probability density function value that said object is on said second avoidance course and determining that said object is likely to be between first and second collision courses when said first and second ratios are greater than a certain value, said certain value being 1-α

)/β

, where α

is the probability of a missed determination of an imminent collision and β

is the probability of a false determination of an imminent collision.

3. The process of claim 2, wherein said certain value is about 80000.

4. The process of claim 1, wherein said determination is made based on a position data of a single observation.

5. The process of claim 1, wherein step (b) is performed using position data from a sequence of observations.

6. The process of claim 1, wherein step (b) is performed before determining target parameters of said object.

7. A process of determining if a collision between a vehicle and an object is imminent, said vehicle having a sensing system for obtaining position data on at least one object within a field of detection, said process comprising:

(a) obtaining position data for one or more observations of said object;

(b) determining first and second collision values, said first collision value being based on a first probability density function of said object being on a first collision course based on said position data, and said second collision value being based on a second probability density function of said object being on a second collision course based on said position data, said first and second collision courses defining boundaries of a collision zone;

(c) determining first and second avoidance values, said first avoidance value being based on a first probability density function that said object is on a first avoidance course based on position data, and said second avoidance value being based on a second probability density function that said object is on a second avoidance course based on said position data, said first avoidance course being the boundary of a safe zone closest to said first collision course and said second avoidance course being the boundary of a second safe zone closest to said second collision course;

(d) determining that a collision is imminent if the ratio of said first collision value to said first avoidance value is above a first certain number, and if the ratio of said second collision value to said second avoidance is above a second certain number; and

(e) providing information to facilitate deployment of collision mitigation measures if a collision is determined to be imminent.

8. The process of claim 7, wherein said first and second certain numbers are approximately (1-α

)/β

, where α

is the probability of a missed determination of an imminent collision and β

is the probability of a false determination of an imminent collision.

9. The process of claim 8, wherein said certain value is about 80000.

10. The process of claim 9, wherein the probability that said object is actually within a collision zone is greater than about 80% and the probability that the object is actually within a safe zone is less than about 0.001%.

11. The process of claim 7, wherein position data is sequential data obtained for two or more sequential observations.

12. The process of claim 11, wherein determining whether said observed object is an actual object is determined using a retrospective-detection algorithm.

13. The process of claim 11, wherein the likelihood that said observed object is an actual object is determined by comparing said sequential position data to a pattern.

14. The process of claim 13, wherein said pattern is a straight line representing the trajectory of said observed object relative to said vehicle.

15. The process of claim 7, wherein said collision zone and said safe zones are determined independently of the position of said object.

16. The process of claim 15, wherein said collision zone is a course in front of said vehicle and said first safe zone is a certain distance to the left of said collision zone and said second safe zone is a certain distance to the right of said collision zone.

17. The process of claim 7, wherein said collision zone and said safe zones are determined based on the position of said object.

18. The process of claim 17, wherein said collision zone for an object is defined on a first side by said first collision course extending between said object and the rear most portion of said vehicle, and on a second side by said second collision course extending between said object and the forward most portion of said vehicle, and said first safe zone is defined on the first side by said first avoidance course extending between said object and a position a certain distance rearward of said vehicle and said second safe zone is defined on the second side by said second avoidance course extending between said object and a position a certain distance forward of said vehicle.

19. The process of claim 7, wherein step (e) comprises determining one or more target parameters for use in pre-collision preparation.

20. The process of claim 19, wherein step (e) comprises determining one or more of the following target parameters:

velocity, direction, and time-to-impact.

21. The process of claim 7, further comprising:

(f) deploying collision mitigation measures if a collision is determined to be imminent.

22. The process of claim 21, wherein said collision mitigation measures comprise at least one of reversible restraints and pre-triggered airbags.

23. The process of claim 7, wherein said first and second collision courses are left and right boundaries, respectively, of said collision zone.

24. The process of claim 7, wherein step (d) involves using a sequential probability ratio test (SPRT).