Unified motion planner for autonomous driving vehicle in avoiding the moving obstacle

• US 9,229,453 B1
• Filed: 08/29/2014
• Issued: 01/05/2016
• Est. Priority Date: 08/29/2014
• Status: Active Grant

• Current Assignee: GM Global Technology Operations LLC (General Motors Company)
• Original Assignee: GM Global Technology Operations LLC (General Motors Company)
• Inventors: Lee, Jin-Woo
• Primary Examiner(s): Tarcza, Thomas
• Assistant Examiner(s): Pipala, Edward
• Application Number:
  US14/473,840
• Time in Patent Office: 494 Days
• US Class Current: 1/1
• CPC Class Codes: B60W 10/184 with wheel brakes
• CPC Class Codes: B60W 10/20 including control of steering systems
• CPC Class Codes: B60W 2050/009 Priority selection
• CPC Class Codes: B60W 2420/403 Image sensing, e.g. optical camera
• CPC Class Codes: B60W 2520/10 Longitudinal speed
• CPC Class Codes: B60W 2552/53 Road markings, e.g. lane marker or crosswalk
• CPC Class Codes: B60W 2554/4044 Direction of movement, e.g. backwards
• CPC Class Codes: B60W 2554/802 Longitudinal distance
• CPC Class Codes: B60W 30/09 Taking automatic action to avoid collision, e.g. braking and steering
• CPC Class Codes: B60W 30/095 Predicting travel path or likelihood of collision
• CPC Class Codes: B60W 30/0953 the prediction being responsive to vehicle dynamic parameters
• CPC Class Codes: B60W 30/12 Lane keeping
• CPC Class Codes: G05D 1/0246 using a video camera in combination with image processing means

Abstract

A system and method for providing path planning and generation in a semi-autonomous or autonomously driven vehicle that provides a steering correction for collision avoidance purposes. The method includes detecting a lane center of a roadway lane that the vehicle is traveling along and determining a lane centering path that directs the vehicle from its current position to the lane center. The method also includes detecting a moving object in front of the vehicle and determining if a collision between the vehicle and the object will occur if the vehicle travels along the lane centering path at the current vehicle speed. The method solves a fifth-order polynomial equation to define a collision avoidance path from the current vehicle position to a waypoint a safe distance from the object and a return path from the waypoint to the lane center that the vehicle is automatically steered along.

Claims

1. A method for providing an automatic steering path for a vehicle, said method comprising:

detecting a lane center of a roadway lane that the vehicle is traveling along;

determining a lane centering path that directs the vehicle from a current vehicle position to the lane center;

automatically steering the vehicle along the lane centering path;

detecting a moving object in front of the vehicle;

determining if a collision between the vehicle and the object will occur if the vehicle travels along the lane centering path at a current vehicle speed;

determining whether vehicle braking alone will prevent the collision with the object if it is determined that the collision will occur;

determining a steering path around the object if it is determined that vehicle braking will not prevent a collision with the object; and

steering the vehicle along the steering path.

2. The method according to claim 1 wherein detecting a lane center of a roadway lane includes using camera measurements of detected roadway markings on the roadway and a map database of roadway geometry and defining the lane center as a third-order polynomial equation based on fusion of the camera measurements and the map database roadway geometry.

3. The method according to claim 1 wherein determining whether vehicle braking alone will prevent the collision includes calculating a required deceleration to avoid the collision and comparing the required deceleration with a maximum capable deceleration of the vehicle.

4. The method according to claim 1 wherein determining if a collision between the vehicle and the object will occur includes defining a collision boundary around the moving object that is based on a speed of the moving object, moving direction of the object, a size of the moving object and a time-to-collision value, and determining if the lane centering path crosses the collision boundary.

5. The method according to claim 4 wherein determining a steering path around the object includes determining a collision avoidance path from the current vehicle position to a waypoint on the collision boundary, said method further comprising steering the vehicle to the waypoint on the collision avoidance path.

6. The method according to claim 5 wherein determining the collision avoidance path includes selecting one of a plurality of possible collision avoidance paths to multiple waypoints on the collision boundary where the closest waypoint to the current vehicle position that is within roadway boundaries is selected as the algorithm avoidance path.

7. The method according to claim 5 wherein determining the collision avoidance path includes solving a fifth-order polynomial equation based on initial vehicle conditions and roadway boundary conditions.

8. The method according to claim 7 wherein the fifth-order polynomial equation is:

y_AP(x)=a_5,APx⁵+a_4,APx⁴+a_3,APx³+a_2,APx²+a_1,APx+a_0,AP,

for time 0<

t<

t_TTC, wherein the initial conditions are defined as;

y_AP(0)=0,

y′

_AP(0)=0,

y″

_AP(0)=2c₂,

and the boundary conditions are defined as;

y_AP(v_xt_TTC)=y_lane(v_xt_TTC)+D_LC,

y′

_AP(v_xt_TTC)=y′

_lane(v_xt_TTC),

y″

_AP(v_xt_TTC)=y″

_lane(v_xt_TTC),

where y_AP is vehicle lateral position on the collision avoidance path, y_lane is vehicle lateral position on the lane centering path, a_0,AP, a_1,AP, a_2,AP, a_3,AP, a_4,AP and a_5,AP are unknown coefficients at to be determined using the initial conditions and the boundary conditions, t_TTC is the time-to-collision value, v_x is vehicle longitudinal speed, and D_LC is a lateral deviation from the lane center.

9. The method according to claim 5 wherein detecting a moving object in front of the vehicle includes determining whether the moving object is moving in a transverse direction or a lateral direction relative to the lane centering path.

10. The method according to claim 9 wherein determining a steering path around the object includes determining a return path from the waypoint to the lane center if the object is moving in a transverse direction.

11. The method according to claim 10 wherein determining a return path includes determining the return path by solving a fifth-order polynomial equation based on initial conditions and roadway boundary conditions.

12. The method according to claim 11 wherein the fifth-order polynomial equation is:

y_RP(x)=a_5,RPx⁵+a_4,RPx⁴+a_3,RPx³+a_2,RPx²+a_1,RPx+a_0,RP,

for time t_TTC<

t<

t_TTR, wherein the initial conditions are defined as;

y_RP(v_xt_TTC)=y_AP(v_xt_TTC),

y′

_RP(v_xt_TTC)=y′

_AP(v_xt_TTC),

y″

_RP(v_xt_TTC)=y″

_AP(v_xt_TTC),

and the boundary conditions are defined as;

y_RP(v_xt_TTR)=y_lane(v_xt_TTR),

y′

_RP(v_xt_TTR)=y′

_lane(v_xt_TTR),

y″

_RP(v_xt_TTR)=y″

_lane(v_xt_TTR),

where y_AP is vehicle lateral position on the collision avoidance path, y_RP is vehicle lateral position on the return path, y_lane is vehicle lateral position on the lane centering path, a_0,RP, a_1,RP, a_2,RP, a_3,RP, a_4,RP and a_5,RP are unknown coefficients and to be determined using the initial conditions and the boundary conditions, t_TTC is the time-to-collision value, t_TTR is a time-to-return value, and v_x is vehicle longitudinal speed.

13. The method according to claim 10 wherein determining a vehicle steering path includes determining a cruise path between the collision avoidance path and the return path if the object is moving in a longitudinal direction relative to the vehicle or if the object is too long to be avoided only by the collision avoidance path and the return path.

14. The method according to claim 13 wherein determining the cruise path includes solving a fifth-order polynomial equation based on initial conditions and roadway boundary conditions.

15. The method according to claim 14 wherein the fifth-order polynomial equation is:

y_CP(x)=a_5,CPx⁵+a_4,CPx⁴+a_3,CPx³+a_2,CPx²+a_1,CPx+a_0,CP,

for time t_TTC<

t<

t_TTS, wherein the initial conditions are defined as;

y_CP(v_xt_TTC)=y_AP(v_xt_TTC),

y′

_CP(v_xt_TTC)=y′

_AP(v_xt_TTC),

y″

_CP(v_xt_TTC)=y″

_AP(v_xt_TTC),

and the boundary conditions are defined as;

y_CP(v_xt_TTS)=y_lane(v_xt_TTS)+D_LC,

y′

_CP(v_xt_TTS)=y′

_lane(v_xt_TTS),

y″

_CP(v_xt_TTS)=y″

_lane(v_xt_TTS),

where y_AP is vehicle lateral position on the collision avoidance path, y_CP is vehicle lateral position on the cruise path, y_lane is vehicle lateral position on the lane centering path, a_0,CP, a_1,CP, a_2,CP, a_3,CP, a_4,CP and a_5,CP are unknown coefficients and to be determined using the initial conditions and the boundary conditions, t_TTC is the time-to-collision value, t_TTS is a time-to-cruise value, v_x is vehicle longitudinal speed, and D_LC is a deviation from the lane center.

16. The method according to claim 15 wherein a length of the cruise path is defined by the time-to-cruise value that is determined from the speed of the object and its direction as:

17. A method for providing an automatic steering path for a vehicle, said method comprising:

detecting a lane center of a roadway lane that the vehicle is traveling along;

determining a lane centering path that directs the vehicle from a current vehicle position to the lane center;

automatically steering the vehicle along the lane centering path;

detecting a moving object in front of the vehicle where the moving object is moving in a transverse direction relative to the lane centering path;

determining if a collision between the vehicle and the object will occur if the vehicle travels along the lane centering path at a current vehicle speed;

determining a collision avoidance path from the current vehicle position to a safe waypoint adjacent to the moving object, wherein determining the collision avoidance path includes solving a fifth-order polynomial equation;

steering the vehicle from the current vehicle position to the waypoint on the collision avoidance path;

determining a return path from the waypoint to the lane center, wherein determining a return path includes solving a fifth-order polynomial equation; and

steering the vehicle from the waypoint on the return path.

18. The method according to claim 17 wherein detecting a lane center of a roadway lane includes using camera measurements of detected roadway markings on the roadway and map database of roadway geometry and defining the lane center as a third-order polynomial equation based on fusion of the camera measurements and the map database roadway geometry.

19. A method for providing an automatic steering path for a vehicle, said method comprising:

detecting a lane center of a roadway lane that the vehicle is traveling along;

determining a lane centering path that directs the vehicle from a current vehicle position to the lane center;

automatically steering the vehicle along the lane centering path;

detecting a moving object in front of the vehicle where the moving object is moving in a longitudinal direction relative to the lane centering path;

determining if a collision between the vehicle and the object will occur if the vehicle travels along the lane centering path at a current vehicle speed;

determining a collision avoidance path from the current vehicle position to a safe waypoint adjacent to the moving object, wherein determining the collision avoidance path includes solving a fifth-order polynomial equation;

steering the vehicle from the current vehicle position to the waypoint on the collision avoidance path;

determining a cruise path from the waypoint to an avoidance location beyond the object, wherein determining the cruise path includes solving a fifth-order polynomial equation;

steering the vehicle from the waypoint to the avoidance location on the cruise path;

determining a return path from the avoidance location to the lane center, wherein determining a return path includes solving a fifth-order polynomial equation; and

steering the vehicle from the avoidance location on the return path.

20. The method according to claim 19 wherein detecting a lane center of a roadway lane includes using camera measurements of detected roadway markings on the roadway and map database of roadway geometry and defining the lane center as a third-order polynomial equation based on fusion of the camera measurements and the map database roadway geometry.