System for and method of detecting a collision and predicting a vehicle path

• US 20070043502A1
• Filed: 08/18/2005
• Published: 02/22/2007
• Est. Priority Date: 08/18/2005
• Status: Active Grant

• Current Assignee: GM Global Technology Operations LLC (General Motors Company)
• Original Assignee: GM Global Technology Operations Incorporated (General Motors Company)
• Inventors: Kellum, Carroll, Mudalige, Upali
• Granted Patent: US 7,729,857 B2
• Time in Patent Office: Days
• US Class Current: 701/207
• CPC Class Codes: B60T 7/22 initiated by contact of vehicle, e.g. bumper, with an external object, e.g. another vehicle , or by means of contactless obstacle detectors mounted on the vehicle
• CPC Class Codes: G08G 1/163 involving continuous checking

Abstract

A collision detection and path prediction system (10) adapted for use with a traveling host vehicle (12) having an operator, includes a locator device (20) configured to determine the current position coordinates, and pluralities of trail and immediate dynamic path coordinates of the vehicle (12) and a communicatively coupled traveling remote vehicle (16). The system further includes a preferred controller (36) configured to predict a collision between the two vehicles (12,16) from the coordinates, and determine a plurality of projected path coordinates for the host vehicle (12) relative to the remote vehicle (16) trail coordinates.

Claims

1. A path prediction system (10) adapted for use with a traveling host vehicle (12) spaced from and communicatively coupled to at least one traveling remote vehicle (16), said system comprising:

a locator device (20) configured to determine and store for at least a period current position coordinates, and a plurality of trail coordinates for each of said host and at least one remote vehicle (12,16); and

a controller (36) communicatively coupled to the device (20), and configured to autonomously determine a heading, and a path for each of said host and remote vehicles (12,16), based on the current position coordinates, and pluralities of trail coordinates,

said controller (36) being further configured to determine a vehicle speed for each of said host and remote vehicles (12,16), and generate a signal or command, where the relative speeds, position coordinates, and headings project concurrently intersecting paths.

2. The system as claimed in claim 1,

said locator device (20) being further configured to determine and store for at least a period relative current position and trail coordinates between the host and said at least one remote vehicle (12,16), so as to present a relative coordinate system.

3. The system as claimed in claim 1,

said controller (36) being configured to further determine a projected path for the host vehicle (12), and generate a second signal or command, when a host vehicle current position coordinate deviates from the projected path.

4. The system as claimed in claim 3,

said controller (36) being further configured to create a plurality of differing signals, wherein each signal indicates a different degree of deviation from the projected path.

5. The system as claimed in claim 4,

said signal being further configured to direct the host vehicle (12) towards the projected path.

6. The system as claimed in claim 1,

said device (20) including a map database (32) having at least one record (32a), wherein said record (32a) presents a plurality of global positioning points,

said device (20) being configured to match the current position coordinate, and at least a portion of the remote vehicle trail coordinates with corresponding points on the record (32a), and display the record,

said controller (36) being communicatively coupled to the database (32), and configured to determine a plurality of projected path coordinates for the host vehicle (12), and match the projected path coordinates with corresponding points on the record (32a).

7. A method of predicting the path of a traveling host vehicle (12) relative to at least one traveling remote lead vehicle (16), by a controller (36), said method comprising the steps of:

a) determining and storing a plurality of trail coordinates, a current position coordinate, and heading, for at least one traveling remote vehicle (16);

b) determining a current position coordinate and heading, for the host vehicle (12);

c) comparing the current position coordinates, and headings, for the host vehicle (12) and said at least one traveling remote vehicle (16), to determine at least one traveling remote lead vehicle, wherein the lead vehicle is generally ahead of and traveling in the same general direction with the host vehicle (12); and

d) determining a plurality of projected path coordinates for the host vehicle (12), wherein each path coordinate is matched with and congruently related to a corresponding trail coordinate of the lead vehicle.

8. The method as claimed in claim 7,

steps a), b) and c) further including the steps of determining a vehicle speed V for each of said host and remote vehicles (12,16), determining and storing for each, a plurality of IDP coordinates based on the corresponding plurality of trail coordinates, current position coordinates, heading, and speed, and comparing the current position coordinates, headings, speeds, and IDP coordinates for the host vehicle (12) and said at least one traveling remote vehicle (16), to determine at least one traveling remote lead vehicle.

9. The method as claimed in claim 8,

steps a) and b) further including the steps of determining a vehicle acceleration f, yaw rate, and steering wheel angle for each of said host and remote vehicles (12,16), and determining the plurality of IDP coordinates, such that the IDP coordinates present a length L based on the V, f, yaw rate, and/or steering wheel angle of the corresponding vehicle.

10. The method as claimed in claim 8,

steps a) and b) further including the steps of determining a vehicle acceleration f for each of said host and remote vehicles (12,16), and determining for each the IDP coordinates, such that the IDP coordinates present an IDP length L, in accordance with the following equation:

L=2(V+f).

11. The method as claimed in claim 7,

step c) further including the steps of determining the direction of travel for said at least one remote vehicle (16) relative to the host vehicle (12) based on the relative heading between the vehicles (12,16).

12. The method as claimed in claim 11,

step c) further including the steps of determining the direction of travel for said at least one remote vehicle (16) relative to the host vehicle (12), according to the following inequality:

Cos θ

_h Cos θ

_s+Sin θ

_h Sin θ

_s>

0.5,

wherein θ

_h is the host vehicle heading, θ

_s is the remote vehicle heading, and a true value for the inequality yields the same relative direction of travel.

13. The method as claimed in claim 7,

step c) further including the steps of determining the ahead or behind condition of said at least one remote vehicle (16) relative to the host vehicle (12) based on the headings and current position coordinates of the vehicles (12,16).

14. The method as claimed in claim 13,

step c) further including the steps of determining the condition of said at least one remote vehicle (16) relative to the host vehicle (12), according to the following inequality:

(x_s−

x_h)Cos θ

_h+(y_s−

y_h)Sin θ

_h<

0,

wherein θ

_h is the host vehicle heading, x_s,y_s are coordinates of the remote vehicle (16), x_h,y_h are coordinates of the host vehicle (12), and a true value for the inequality occurs when the remote vehicle (16) is behind the host vehicle (12).

15. The method as claimed in claim 7,

step c) further including the steps of categorizing said at least one remote vehicle (16) into one of a plurality of categories, based on the vehicle heading and current position coordinates, wherein said categories are selected from a group consisting essentially of different roadway height, oncoming, intersecting left, intersecting right, ahead host lane, ahead adjacent lane, ahead offset lane, behind host lane, behind adjacent lane, and behind offset lane.

16. The method as claimed in claim 15, wherein the relative lane position for each remote vehicle (16) is determined by analyzing a direction vector sequentially taken from a parametrically interior host vehicle point a, to a front most host vehicle point b, and to a parametrically interior remote vehicle trail point S, so that only a counter-clockwise vector is positive, and the remote vehicle (16) is located in a leftward lane when the following inequality is true:

(b_x−

a_x)(s_y−

a_y)−

(s_x−

a_x)(b_y−

a_y)>

0,

and, a_x,a_y are coordinates of the point a, b_x,b_y are coordinates of the point b, and s_x,x_y are coordinates of the point S.

17. The method as claimed in claim 15, wherein the relative lane position for said at least one remote vehicle (16) is determined by computing lateral offsets between trail and current position coordinates, for the host vehicle (12) and said at least one remote vehicle (16), and comparing the offsets to a plurality of thresholds.

18. The method as claimed in claim 15, wherein the remote vehicle (16) is categorized in the ahead host lane, and the plurality of projected path coordinates for the host vehicle (12) is determined such that each of the path coordinates is congruent to a corresponding remote vehicle trail coordinate.

19. The method as claimed in claim 15, wherein the remote vehicle (16) is categorized in the ahead adjacent or offset lane, and the plurality of projected path coordinates for the host vehicle (12) is determined such that each of the path coordinates is spaced at a generally constant lateral offset, as measured perpendicularly from a corresponding remote vehicle trail coordinate.

20. The method as claimed in claim 19, wherein the lead vehicle is further configured to detect a lane change and transmit the coordinates and time of the lane change to the host vehicle (12), and the host vehicle (12) is further configured to modify the categorization or lateral offset based on the remote vehicle lane change information.

21. The method as claimed in claim 19, wherein a plurality of remote vehicles (16) are categorized in the ahead adjacent and offset lanes, initial pluralities of projected path coordinates for the host vehicle (12) are determined relative to each of said plurality of remote vehicles (16) and compared, and a final plurality of projected path coordinates is determined from the longest of said initial pluralities of path coordinates.

22. A method of predicting the paths of a traveling host vehicle (12) and at least one traveling remote vehicle (16), and avoiding a future concurrent intersection of the paths, by a controller (36), said method comprising the steps of:

a) determining and storing a plurality of the path coordinates for the host vehicle (12);

b) determining and storing a plurality of the path coordinates for said at least one remote vehicle (16);

c) electronically comparing the pluralities of path coordinates to detect a future concurrent intersection of path coordinates; and

d) generating electrical signals for causing the path coordinates of either the host or remote vehicle to be modified, so as to eliminate the future concurrent intersection of the paths.

23. The method as claimed in claim 22,

steps a), b), and d) further including the steps of determining a host vehicle speed, a remote vehicle speed, and causing the host vehicle (12) to decelerate or accelerate.

24. The method as claimed in claim 22,

steps a), b) and c) further including the steps of determining a path polygon, for each of said host and remote vehicles (12,16), based on the corresponding plurality of path coordinates, and vehicle width, and detecting a concurrent overlap area between the path polygons.

25. The method as claimed in claim 22,

step c) further including the steps of comparing the pluralities of path coordinates to detect a future generally concurrent intersection of path coordinates.

26. The method as claimed in claim 22,

steps a), b) and c) further including the steps of determining and storing for each of said host and remote vehicles (12,16) a plurality of trail coordinates, and a current position coordinate, and comparing the pluralities of trail coordinates, and current position coordinates.

27. The method as claimed in claim 26,

steps a), b) and c) further including the steps of determining a path polygon, and a trail polygon for each of said host and remote vehicles (12,16), based on the corresponding plurality of path or trail coordinates, and the vehicle width, and detecting an overlap area between one of the host path and trail polygons and one of the remote path and trail polygons.

28. The method as claimed in claim 27,

step c) further including the steps of determining a temporally increasing or decreasing area.

29. A computer program implementing the method of claim 22 when executed by an electronic device associated with a host vehicle (12) traveling a distance from at least one traveling remote vehicle (16) or object.