Motion planning for non-standard driving behaviors

Patent metadata

Field	Value
Publication number	US-12240491-B2
Application number	US-202218062310-A
Country	US
Kind code	B2
Filing date	Dec 6, 2022
Priority date	Dec 6, 2022
Publication date	Mar 4, 2025
Grant date	Mar 4, 2025

Abstract

A system for motion planning for a vehicle includes at least one vehicle sensor for determining information about an environment surrounding the vehicle and a controller in electrical communication with the at least one vehicle sensor. The controller is programmed to perform a plurality of measurements of a remote vehicle using the at least one vehicle sensor. The plurality of measurements includes at least a plurality of position measurements of the remote vehicle. The controller is further programmed to determine a risk score for each of a plurality of location cells in an environment surrounding the remote vehicle based at least in part on the plurality of measurements of the remote vehicle. The controller is further programmed to adjust a planned path of the vehicle based at least in part on the risk score of each of the plurality of location cells in the environment surrounding the remote vehicle.

First claim

What is claimed is:

1. A system for motion planning for a vehicle, the system comprising:
at least one vehicle sensor for determining information about an environment surrounding the vehicle; and
a controller in electrical communication with the at least one vehicle sensor, wherein the controller is programmed to:
perform a plurality of measurements of a remote vehicle using the at least one vehicle sensor, wherein the plurality of measurements includes at least a plurality of position measurements of the remote vehicle relative to the vehicle;
determine a risk score for each of a plurality of location cells in an environment surrounding the remote vehicle based at least in part on the plurality of measurements of the remote vehicle; and
adjust a planned path of the vehicle based at least in part on the risk score of each of the plurality of location cells in the environment surrounding the remote vehicle, wherein to adjust the planned path of the vehicle, the controller is further programmed to:
determine a behavior class of the remote vehicle based at least in part on the plurality of measurements of the remote vehicle, wherein the behavior class includes an intentional behavior class and an unintentional behavior class; and
adjust the planned path of the vehicle using a path determination algorithm based at least in part on the behavior class, wherein to adjust the planned path of the vehicle, the controller is further programmed to:
determine a behavior type of the remote vehicle in response to determining that the behavior class of the remote vehicle is the intentional behavior class, wherein the behavior type includes at least one of a tailgating behavior type, a road rage behavior type, and a wrong direction behavior type;
execute a tailgating supervisory action to adjust the planned path of the vehicle in response to determining that the behavior type of the remote vehicle is the tailgating behavior type;
execute a road rage supervisory action to adjust the planned path of the vehicle in response to determining that the behavior type of the remote vehicle is the road rage behavior type; and
execute a wrong direction supervisory action to adjust the planned path of the vehicle in response to determining that the behavior type of the remote vehicle is the wrong direction behavior type.

2. The system of claim 1 , wherein to determine the risk score for each of the plurality of location cells in the environment surrounding the remote vehicle, the controller is further programmed to:
determine the risk score for each of the plurality of location cells using a risk score machine learning model, wherein an input for the risk score machine learning model includes the plurality of measurements of the remote vehicle, and wherein an output of the risk score machine learning model is the risk score for each of the plurality of location cells.

3. The system of claim 1 , wherein to execute the tailgating supervisory action, the controller is further programmed to:
determine a maximum allowed speed based at least in part on a speed limit of a roadway upon which the vehicle is traveling;
compare a speed of the vehicle to the maximum allowed speed;
increase the speed of the vehicle in response to determining that the speed of the vehicle is less than the maximum allowed speed;
maintain the speed of the vehicle in response to determining that the speed of the vehicle is greater than or equal to the maximum allowed speed;
identify a state of a right adjacent lane of travel, wherein the state of the right adjacent lane of travel includes an occupied state and an unoccupied state;
move the vehicle into the right adjacent lane of travel in response to determining that the state of the right adjacent lane of travel is the unoccupied state;
identify a state of a left adjacent lane of travel, wherein the state of the left adjacent lane of travel includes an occupied state and an unoccupied state;
move the vehicle into the left adjacent lane of travel in response to determining that the state of the left adjacent lane of travel is the unoccupied state;
determine a chase time in response to determining that the state of the right adjacent lane of travel is the occupied state and the state of the left adjacent lane of travel is the occupied state;
compare the chase time to a predetermined chase time threshold;
notify an occupant of the vehicle and move the vehicle into a shoulder of the roadway in response to determining that the chase time is greater than or equal to the predetermined chase time threshold; and
transmit information to a remote vehicle using a vehicle communication system in response to determining that the chase time is greater than or equal to the predetermined chase time threshold.

4. The system of claim 1 , wherein to execute the road rage supervisory action, the controller is further programmed to:
notify an occupant of the vehicle;
determine a lane state of the remote vehicle, wherein the lane state includes a same lane state and an adjacent lane state;
move the vehicle into an adjacent lane in response to determining that the lane state of the remote vehicle is the same lane state;
decrease a speed of the vehicle in response to determining that the lane state of the remote vehicle is the adjacent lane state;
identify a relative location of the vehicle to the remote vehicle, wherein the relative location of the vehicle includes a leading relative location and a following relative location;
take a first evasive action in response to determining that the relative location of the vehicle is the leading relative location; and
take a second evasive action in response to determining that an occupant of the remote vehicle has exited the remote vehicle.

5. The system of claim 1 , wherein to execute the wrong direction supervisory action, the controller is further programmed to:
detect the remote vehicle using a vehicle communication system;
adjust the planned path of the vehicle to exit a roadway upon which the remote vehicle is traveling in response to detecting the remote vehicle using the vehicle communication system;
detect the remote vehicle using the at least one vehicle sensor;
identify a predicted path of the remote vehicle in response to detecting the remote vehicle using the at least one vehicle sensor, wherein the predicted path includes a collision path and a non-collision path;
move the vehicle to a shoulder of a roadway upon which the vehicle is traveling in response to determining that the predicted path of the remote vehicle is the collision path; and
take a third evasive action in response to determining that the predicted path of the remote vehicle is the non-collision path.

6. The system of claim 1 , wherein the path determination algorithm is configured to adjust the planned path of the vehicle to minimize the risk score of each of the plurality of location cells entered by the vehicle and maximize a distance between the vehicle and the remote vehicle.

7. The system of claim 6 , wherein the path determination algorithm is a reinforcement learning algorithm, and wherein the reinforcement learning algorithm is trained based at least in part on a distance between the vehicle and the remote vehicle.

8. The system of claim 7 , wherein the reinforcement learning algorithm is trained using a simulated environment, wherein the simulated environment includes a simulated host vehicle and a simulated remote vehicle, wherein the simulated environment further includes a plurality of simulated location cells, each of the plurality of simulated location cells having a simulated risk score, wherein the reinforcement learning algorithm receives a first reward proportional to the distance between the simulated hos

Assignees

• Gm Global Tech Operations Llc

Inventors

• Esfahani Alireza Esna Ashari
• Mudalige Upali P

Classifications

• B60W30/0956

  the prediction being responsive to traffic or environmental parameters · CPC title

• B60W2554/4046

  Behavior, e.g. aggressive or erratic · CPC title

• B60W50/14

  Means for informing the driver, warning the driver or prompting a driver intervention · CPC title

• B60W2554/80

  Spatial relation or speed relative to objects · CPC title

• B60W30/18163

  Lane change; Overtaking manoeuvres · CPC title