Granted Patent B2

US 12,397,828 · App. 18/453,489 · Granted Aug 26, 2025

Inventors: Michael Cui (Winnetka, CA); Hyukseong Kwon (Thousand Oaks, CA); Rodolfo Valiente Romero (Calabasas, CA); Andrew Howe (Malibu, CA); Alexander Waagen (Malibu, CA); Alexei Kopylov (Malibu, CA); Marcus James Huber (Saline, MI); Alireza Esna Ashari Esfahani (Daly City, CA)

Assignee: GM GLOBAL TECHNOLOGY OPERATIONS LLC

B60W60/0016B60W30/09B60W30/0956B60W60/00182B60W60/00272B60W60/00274B60W60/00276

Abstract

Claims (194)

1. A method for path planning for a vehicle, the method comprising:

determining a predicted trajectory of a remote vehicle, wherein the predicted trajectory of the remote vehicle includes a plurality of predicted trajectory nodes;

determining a plurality of possible trajectories for the vehicle, wherein the plurality of possible trajectories includes a plurality of possible trajectory nodes;

performing a closeness check of the plurality of possible trajectories, wherein the closeness check includes removing a first trajectory from the plurality of possible trajectories in response to determining that the first trajectory includes one or more of the plurality of possible trajectory nodes which are within a predetermined minimum stopping distance of one or more of the plurality of predicted trajectory nodes;

determining one or more evaluation metrics of each of the plurality of possible trajectory nodes based at least in part on a weather condition in an environment surrounding the vehicle, wherein the one or more evaluation metrics includes a collision potential score, a maneuverability score, and a traffic violation score;

selecting an optimal trajectory for the vehicle from the plurality of possible trajectories based at least in part on the one or more evaluation metrics of each of the plurality of possible trajectory nodes; and

performing a first action based at least in part on the optimal trajectory.

2. The method of claim 1 , wherein determining the collision potential score for each of the plurality of possible trajectory nodes further comprises:

determining a plurality of uncertainties, wherein the plurality of uncertainties includes a plurality of predicted trajectory uncertainties and a plurality of possible trajectory uncertainties, wherein each of the plurality of predicted trajectory uncertainties corresponds to one of the plurality of predicted trajectory nodes, and wherein each of the plurality of possible trajectory uncertainties corresponds to one of the plurality of possible trajectory nodes;

determining a plurality of biased predicted trajectory uncertainties and a plurality of biased possible trajectory uncertainties by applying a bias to each of the plurality of predicted trajectory uncertainties and each of the plurality of possible trajectory uncertainties, wherein the bias is based at least in part on the weather condition in the environment surrounding the vehicle; and

determining the collision potential score for each of the plurality of possible trajectory nodes based at least in part on the plurality of biased predicted trajectory uncertainties and the plurality of biased possible trajectory uncertainties.

3. The method of claim 2 , wherein determining the plurality of uncertainties further comprises:

determining the plurality of uncertainties based on at least one of: the weather condition in the environment surrounding the vehicle and a localization error of one or more of a plurality of vehicle sensors.

4. The method of claim 2 , wherein determining the plurality of biased predicted trajectory uncertainties and the plurality of biased possible trajectory uncertainties further comprises:

determining a longitudinal bias for each of the plurality of possible trajectory nodes and each of the plurality of predicted trajectory nodes;

determining a lateral bias for each of the plurality of possible trajectory nodes and each of the plurality of predicted trajectory nodes, wherein the lateral bias is determined based at least in part on a dynamics-based model and the weather condition in the environment surrounding the vehicle;

applying the longitudinal bias and the lateral bias to each of the plurality of predicted trajectory uncertainties; and

applying the longitudinal bias and the lateral bias to each of the plurality of possible trajectory uncertainties.

5. The method of claim 2 , wherein determining the collision potential score for each of the plurality of possible trajectory nodes further comprises:

determining the collision potential score using a formula:

p

i

=

r

i

⋂

e

i

r

i

⋃

e

i

wherein p i is the collision potential score for an ith node of the plurality of possible trajectory nodes, r i is one of the plurality of biased predicted trajectory uncertainties corresponding to an ith node of the plurality of predicted trajectory nodes, et is one of the plurality of biased possible trajectory uncertainties corresponding to the ith node of the plurality of possible trajectory nodes, ∩ is an intersection operator, and ∪ is a union operator.

6. The method of claim 2 , wherein determining the maneuverability score for each of the plurality of possible trajectory nodes further comprises:

determining an estimated propulsion system torque required to reach each of the plurality of possible trajectory nodes; and

determining the maneuverability score for each of the plurality of possible trajectory nodes based at least in part on the estimated propulsion system torque using a formula:

m

i

=

T

i

T

M

⁢

A

⁢

X

wherein m i is the maneuverability score for an ith node of the plurality of possible trajectory nodes, T i is the estimated propulsion system torque required to reach the ith node of the plurality of possible trajectory nodes, and T MAX is a maximum torque available from a propulsion system of the vehicle.

7. The method of claim 6 , wherein the traffic violation score quantifies a legality of maneuvering the vehicle to each of the plurality of possible trajectory nodes.

8. The method of claim 7 , wherein selecting the optimal trajectory further comprises:

selecting a subset of the plurality of possible trajectory nodes to be the optimal trajectory, wherein the subset of the plurality of possible trajectory nodes is selected to minimize a sum of each of the one or more evaluation metrics of each of the subset of the plurality of possible trajectory nodes.

9. The method of claim 7 , wherein selecting the optimal trajectory further comprises:

selecting a subset of the plurality of possible trajectory nodes to be the optimal trajectory, wherein the subset of the plurality of possible trajectory nodes is selected to minimize an objective function:

c

=

∑

i

=

0

N

N

-

i

N

*

(

p

i

+

v

i

+

m

i

)

wherein c is the objective function, N is a quantity of nodes in the subset of the plurality of possible trajectory nodes, p i is the collision potential score of an ith node of the subset of the plurality of possible trajectory nodes, v i is the traffic violation score of the ith node of the subset of the plurality of possible trajectory nodes, and m i is the maneuverability score of the ith node of the subset of the plurality of possible trajectory nodes, wherein each of the subset of the plurality of possible trajectory nodes is from a same one of the plurality of possible trajectories, and wherein the collision potential score of a first node of the subset of the plurality of possible trajectory nodes and a second node of the subset of the plurality of possible trajectory nodes is less than one.

10. The method of claim 1 , wherein performing the first action further comprises:

adjusting an operation of an automated driving system of the vehicle based at least in part on the optimal trajectory.

11. A system for path planning for a vehicle, the system comprising:

a plurality of vehicle sensors;

an automated driving system; and

a controller in electrical communication with the plurality of vehicle sensors and the automated driving system, wherein the controller is programmed to:

determine a predicted trajectory of a remote vehicle using the plurality of vehicle sensors, wherein the predicted trajectory of the remote vehicle includes a plurality of predicted trajectory nodes;

determine a plurality of possible trajectories for the vehicle using the plurality of vehicle sensors, wherein the plurality of possible trajectories includes a plurality of possible trajectory nodes;

perform a closeness check of the plurality of possible trajectories, wherein the closeness check includes removing a first trajectory from the plurality of possible trajectories in response to determining that the first trajectory includes one or more of the plurality of possible trajectory nodes which are within a predetermined minimum stopping distance of one or more of the plurality of predicted trajectory nodes;

determine one or more evaluation metrics of each of the plurality of possible trajectory nodes based at least in part on a weather condition in an environment surrounding the vehicle, wherein the weather condition is determined using the plurality of vehicle sensors, and wherein the one or more evaluation metrics includes a collision potential score, a maneuverability score, and a traffic violation score;

select an optimal trajectory for the vehicle from the plurality of possible trajectories based at least in part on the one or more evaluation metrics of each of the plurality of possible trajectory nodes; and

adjust an operation of the automated driving system based at least in part on the optimal trajectory.

12. The system of claim 11 , wherein to determine the collision potential score for each of the plurality of possible trajectory nodes, the controller is further programmed to:

determine a plurality of uncertainties, wherein the plurality of uncertainties includes a plurality of predicted trajectory uncertainties and a plurality of possible trajectory uncertainties, wherein each of the plurality of predicted trajectory uncertainties corresponds to one of the plurality of predicted trajectory nodes, and wherein each of the plurality of possible trajectory uncertainties corresponds to one of the plurality of possible trajectory nodes;

determine a plurality of biased predicted trajectory uncertainties and a plurality of biased possible trajectory uncertainties by applying a bias to each of the plurality of predicted trajectory uncertainties and each of the plurality of possible trajectory uncertainties, wherein the bias is based at least in part on the weather condition in the environment surrounding the vehicle; and

determine the collision potential score for each of the plurality of possible trajectory nodes based at least in part on the plurality of biased predicted trajectory uncertainties and the plurality of biased possible trajectory uncertainties.

13. The system of claim 12 , wherein to determine the plurality of biased predicted trajectory uncertainties and the plurality of biased possible trajectory uncertainties, the controller is further programmed to:

determine a longitudinal bias for each of the plurality of possible trajectory nodes and each of the plurality of predicted trajectory nodes;

determine a lateral bias for each of the plurality of possible trajectory nodes and each of the plurality of predicted trajectory nodes, wherein the lateral bias is determined based at least in part on a dynamics-based model and the weather condition in the environment surrounding the vehicle;

apply the longitudinal bias and the lateral bias to each of the plurality of predicted trajectory uncertainties; and

apply the longitudinal bias and the lateral bias to each of the plurality of possible trajectory uncertainties.

14. The system of claim 12 , wherein to determine the collision potential score for each of the plurality of possible trajectory nodes, the controller is further programmed to:

determine the collision potential score using a formula:

p

i

=

r

i

⋂

e

i

r

i

⋃

e

i

wherein p i is the collision potential score for an ith node of the plurality of possible trajectory nodes, r i is one of the plurality of biased predicted trajectory uncertainties corresponding to an ith node of the plurality of predicted trajectory nodes, e i is one of the plurality of biased possible trajectory uncertainties corresponding to the ith node of the plurality of possible trajectory nodes, ∩ is an intersection operator, and ∪ is a union operator.

15. The system of claim 12 , wherein to determine the maneuverability score for each of the plurality of possible trajectory nodes, the controller is further programmed to:

determine an estimated propulsion system torque required to reach each of the plurality of possible trajectory nodes; and

determine the maneuverability score for each of the plurality of possible trajectory nodes based at least in part on the estimated propulsion system torque using a formula:

m

i

=

T

i

T

M

⁢

A

⁢

X

wherein m i is the maneuverability score for an ith node of the plurality of possible trajectory nodes, T i is the estimated propulsion system torque required to reach the ith node of the plurality of possible trajectory nodes, and T MAX is a maximum torque available from a propulsion system of the vehicle.

16. The system of claim 15 , wherein the traffic violation score quantifies a legality of maneuvering the vehicle to each of the plurality of possible trajectory nodes.

17. The system of claim 16 , wherein to select the optimal trajectory, the controller is further programmed to:

select a subset of the plurality of possible trajectory nodes to be the optimal trajectory, wherein the subset of the plurality of possible trajectory nodes is selected to minimize an objective function:

c

=

∑

i

=

0

N

N

-

i

N

*

(

p

i

+

v

i

+

m

i

)

wherein c is the objective function, N is a quantity of nodes in the subset of the plurality of possible trajectory nodes, p i is the collision potential score of an ith node of the subset of the plurality of possible trajectory nodes, v i is the traffic violation score of the ith node of the subset of the plurality of possible trajectory nodes, and m i is the maneuverability score of the ith node of the subset of the plurality of possible trajectory nodes, wherein each of the subset of the plurality of possible trajectory nodes is from a same one of the plurality of possible trajectories, and wherein the collision potential score of a first node of the subset of the plurality of possible trajectory nodes and a second node of the subset of the plurality of possible trajectory nodes is of less than one.

18. A system for path planning for a vehicle, the system comprising:

a plurality of vehicle sensors;

an automated driving system; and

a controller in electrical communication with the plurality of vehicle sensors and the automated driving system, wherein the controller is programmed to:

determine a predicted trajectory of a remote vehicle using the plurality of vehicle sensors, wherein the predicted trajectory of the remote vehicle includes a plurality of predicted trajectory nodes;

determine a plurality of possible trajectories for the vehicle using the plurality of vehicle sensors, wherein the plurality of possible trajectories includes a plurality of possible trajectory nodes;

perform a closeness check of the plurality of possible trajectories, wherein the closeness check includes removing a first trajectory from the plurality of possible trajectories in response to determining that the first trajectory includes one or more of the plurality of possible trajectory nodes which are within a predetermined minimum stopping distance of one or more of the plurality of predicted trajectory nodes;

determine one or more evaluation metrics of each of the plurality of possible trajectory nodes based at least in part on a weather condition in an environment surrounding the vehicle, wherein the one or more evaluation metrics includes a collision potential score, a maneuverability score, and a traffic violation score;

select an optimal trajectory for the vehicle from the plurality of possible trajectories based at least in part on the one or more evaluation metrics of each of the plurality of possible trajectory nodes; and

adjust an operation of the automated driving system, wherein the operation of the automated driving system is adjusted such that the vehicle exits a lane boundary based at least in part on the optimal trajectory.

19. The system of claim 18 , wherein to determine the collision potential score for each of the plurality of possible trajectory nodes, the controller is further programmed to:

determine a plurality of uncertainties, wherein the plurality of uncertainties includes a plurality of predicted trajectory uncertainties and a plurality of possible trajectory uncertainties, wherein each of the plurality of predicted trajectory uncertainties corresponds to one of the plurality of predicted trajectory nodes, and wherein each of the plurality of possible trajectory uncertainties corresponds to one of the plurality of possible trajectory nodes;

determine a plurality of biased predicted trajectory uncertainties and a plurality of biased possible trajectory uncertainties by applying a bias to each of the plurality of predicted trajectory uncertainties and each of the plurality of possible trajectory uncertainties, wherein the bias is based at least in part on the weather condition in the environment surrounding the vehicle; and

determine the collision potential score for each of the plurality of possible trajectory nodes based at least in part on the plurality of biased predicted trajectory uncertainties and a plurality of biased possible trajectory uncertainties.

20. The system of claim 19 , wherein to select the optimal trajectory, the controller is further programmed to:

select a subset of the plurality of possible trajectory nodes to be the optimal trajectory, wherein the subset of the plurality of possible trajectory nodes is selected to minimize an objective function:

c

=

∑

i

=

0

N

N

-

i

N

*

(

p

i

+

v

i

+

m

i

)

wherein c is the objective function, N is a quantity of nodes in the subset of the plurality of possible trajectory nodes, p i is the collision potential score of an ith node of the subset of the plurality of possible trajectory nodes, v i is the traffic violation score of the ith node of the subset of the plurality of possible trajectory nodes, and m i is the maneuverability score of the ith node of the subset of the plurality of possible trajectory nodes, wherein each of the subset of the plurality of possible trajectory nodes is from a same one of the plurality of possible trajectories, and wherein a first node of the subset of the plurality of possible trajectory nodes and a second node of the subset of the plurality of possible trajectory nodes have a collision potential score of less than one.

Assignments (1)

Continuity (1)

References Cited (42)

Cited By (1)

US 12,583,456