Path planning for evasive steering maneuver employing a virtual potential field technique

• US 9,199,668 B2
• Filed: 10/28/2013
• Issued: 12/01/2015
• Est. Priority Date: 10/28/2013
• Status: Active Grant

• Current Assignee: GM Global Technology Operations LLC (General Motors Company)
• Original Assignee: GM Global Technology Operations LLC (General Motors Company)
• Inventors: Zeng, Shuqing, Moshchuk, Nikolai K., Jafari, Rouhollah
• Primary Examiner(s): Jeanglaude, Gertrude Arthur
• Assistant Examiner(s): HUYNH, LUAT T
• Application Number:
  US14/065,279
• Publication Number:
  US 20150120138A1
• Time in Patent Office: 764 Days
• Field of Search: 701/24, 701/41, 701/117, 701/301, 701/532, 701/300
• US Class Current: 1/1
• CPC Class Codes: B62D 15/0265 Automatic obstacle avoidance by steering
• CPC Class Codes: G01S 13/931 of land vehicles
• CPC Class Codes: G01S 15/931 of land vehicles
• CPC Class Codes: G01S 17/931 of land vehicles
• CPC Class Codes: G01S 2013/9318 Controlling the steering
• CPC Class Codes: G01S 2013/93185 Controlling the brakes
• CPC Class Codes: G01S 2013/932 using own vehicle data, e.g. ground speed, steering wheel direction
• CPC Class Codes: G01S 2013/9322 using additional data, e.g. driver condition, road state or weather data
• CPC Class Codes: G01S 2013/9323 Alternative operation using light waves
• CPC Class Codes: G01S 2013/9324 Alternative operation using ultrasonic waves
• CPC Class Codes: G01S 2013/93271 in the front of the vehicles
• CPC Class Codes: G01S 7/4808 Evaluating distance, position or velocity data

Abstract

A system and method for calculating a virtual target path that is used to calculate an evasive steering path around a target object, such as a target vehicle, stopped in front of a subject vehicle. The method includes determining a potential field using a plurality of scan points that is a summation of two-dimensional Gaussian functions, where each Gaussian function has center defined by target object scan points and other object scan points. The method identifies a mesh grid in an X-Y plane where the mesh grid includes mesh grid points at locations where X and Y plane lines cross. The method identifies a local minimum point of the potential field for each X-plane line at each mesh grid point along the Y-plane crossing that X-plane line, where the local minimum point is a curve point. The method then connects the curve points to define the target path.

Claims

1. A method for determining a virtual target path which is used to calculate an evasive steering path for a subject vehicle around a target object in a travel path of the subject vehicle without using lane markings, said method comprising:

a collision avoidance system programmed for;

providing a plurality of scan points identifying detected objects from one or more sensors on the subject vehicle;

separating the plurality of scan points into target object scan points received from the target object and other object scan points received from other objects;

determining a potential field using the plurality of scan points that is a summation of two-dimensional Gaussian functions, where each Gaussian function has a center defined by the target object scan points and the other object scan points;

identifying a mesh grid in an X-Y plane and in a predefined region where the steering path will be located where the mesh grid includes mesh grid points at locations where X and Y plane lines cross, and where X plane lines are in a direction of the travel path of the subject vehicle and Y plane lines are in a perpendicular direction to the X direction;

identifying a potential field value from the potential field at each mesh grid point;

identifying a local minimum point of the potential field for each X plane line at each mesh grid point along a Y plane line crosses that X plane line, where the local minimum point is a curve point and where each X plane line includes a curve point;

calculating a corridor width between the target object and other objects and using the corridor width as a weight coefficient to determine an optimal virtual target path such that a path with a wider corridor width is preferred over a path with a narrow corridor width; and

connecting the curve points to define the virtual target path.

2. The method according to claim 1 wherein determining a potential field that is a summation of the two-dimensional Gaussian functions includes using:

U(X,Y)=Σ

_i=1ⁿU_i(X,Y)+Σ

_j=1^mU_i(X,Y),

where U(X, Y) is the potential field, n and m are the number of target object points a_i, i=1, 2, . . . , n, and the object points, b_j, j=1, 2, . . . , m, respectively, and;

3. The method according to claim 1 wherein identifying a local minimum point of the potential field includes using:

4. The method according to claim 3 wherein the identifying a local minimum point of the potential field also includes using:

5. The method according to claim 4 wherein identifying a local minimum point of the potential field also includes using:

6. The method according to claim 1 wherein connecting the curve points to define the virtual target path includes fitting a quadratic polynomial function through the points using a least-square process.

7. The method according to claim 1 further comprising reducing the complexity of the plurality of scan points using a clustering algorithm before the scan points are separated into the target object scan points and the other object scan points.

8. The method according to claim 1 further comprising ignoring scan points that are outside of a predetermined safe distance from the subject vehicle.

9. The method according to claim 1 further comprising detecting a target vehicle as the target object from the other objects.

10. The method according to claim 1 wherein calculating a corridor width between the target object and the other objects includes aborting an evasive steering maneuver if the corridor width is less than a predetermined threshold width.

11. The method according to claim 1 wherein the at least one sensor is at least one LiDAR sensor.

12. A method for determining a virtual target path that is used to calculate an evasive steering path for a subject vehicle around a target vehicle in a travel path of the subject vehicle without using lane markings, said method comprising:

a collision avoidance system programmed for;

providing a plurality of scan points identifying detected objects from one or more LiDAR sensors on the subject vehicle;

separating the plurality of scan points into target vehicle scan points received from the target vehicle and other object scan points received from other objects;

determining a potential field using the plurality of scan points that is a summation of two-dimensional Gaussian functions, where each Gaussian function has a center defined by the target vehicle scan points and the other object scan points;

identifying a mesh grid in an X-Y plane and in a predefined region where the steering path will be located where the mesh grid includes mesh grid points at locations where X and Y plane lines cross, and where X plane lines are in a direction of the travel path of the subject vehicle and Y plane lines are in a perpendicular direction to the X direction;

identifying a potential field value from the potential field at each mesh grid point;

identifying a local minimum point of the potential field for each X plane line at each mesh grid point along a Y plane line that crosses the X plane line, where the local minimum point is a curve point and where each X plane line includes a curve point;

calculating a corridor width between the target object and other objects and using the corridor width as a weight coefficient to determine an optimal virtual target path such that a path with a wider corridor width is preferred over a path with a narrow corridor width; and

connecting the curve points to define the virtual target path by fitting a quadratic polynomial function through the points using a least-square process.

13. The method according to claim 12 wherein determining a potential field that is a summation of the two-dimensional Gaussian functions includes using:

U(X,Y)=Σ

_i=1ⁿU_i(X,Y)+Σ

_j=1^mU_j(X,Y),

where U(X, Y) is the potential field, n and m are the number of target object points a_i, i=1, 2, . . . , n, and the object points, b_j, j=1, 2, . . . , m, respectively, and;

14. The method according to claim 12 wherein identifying a local minimum point of the potential field includes using:

15. The method according to claim 14 wherein the identifying a local minimum point of the potential field also includes using:

16. The method according to claim 15 wherein identifying a local minimum point of the potential field also includes using:

17. The method according to claim 12 further comprising ignoring scan points that are outside of a predetermined safe distance from the subject vehicle.

18. The method according to claim 12 further comprising detecting a target vehicle as the target object from the other objects.

19. The method according to claim 12 wherein calculating a corridor width between the target object and the other objects includes aborting an evasive steering maneuver if the corridor width is less than a predetermined threshold width.