Granted Patent B2

US 12,325,445 · App. 18/101,169 · Granted Jun 10, 2025

Inventors: Hyun Myung (Daejeon, KR); Euigon Jung (Daejeon, KR); Wonho Song (Daejeon, KR)

Assignee: Korea Advanced Institute of Science and Technology

B60W60/0011B60W60/0015B60W2554/20B60W2554/40B60W2556/40

Abstract

Claims (28)

1. A high-definition (HD) map-based local path planning method for dynamic and static object avoidance performed by a computer device, the method comprising:

verifying a position on a global path by parameterizing the global path using an arc length;

generating a path candidate group based on the global path; and

selecting an optimal path as a path having a smallest cost function related to a global path following from the path candidate group,

wherein the cost function includes a global-path-following cost for selecting a path candidate group having a smallest lateral offset from the global path,

wherein the generating of the path candidate group comprises determining, as an obstacle, an object that is not mapped to the HD map among objects determined through a sensor, computing a difference in the arc length from a current vehicle to each obstacle, and determining an arc length value of a path depending on presence or absence of the obstacle.

2. The method of claim 1 , further comprising:

determining a velocity of a vehicle for the selected optimal path.

3. The method of claim 1 , wherein the verifying of the position on the global path comprises interpolating waypoints of the global path generated based on the HD map into a cubic spline curve.

4. The method of claim 1 , wherein the verifying of the position on the global path comprises expressing each waypoint corresponding to the global path as coordinates by parameterizing a position of a vehicle in a road environment with respect to the arc length.

5. The method of claim 1 , wherein the verifying of the position on the global path comprises verifying the position on the global path by expressing a position of a vehicle as an arc length and a lateral offset with respect to the global path.

6. The method of claim 1 , wherein the generating of the path candidate group comprises generating the path candidate group by applying a lateral offset based on the global path.

7. The method of claim 1 , wherein the generating of the path candidate group comprises applying each path candidate group for vehicle control by converting from a Cartesian coordinate system to a curvilinear coordinate system.

8. The method of claim 1 , wherein the cost function includes a static object cost, a smoothness cost, a global-path-following cost, and a dynamic object cost, and is determined by adjusting an importance of each cost.

9. The method of claim 1 , wherein the selecting of the optimal path comprises allowing driving by reflecting a cost function for a dynamic object and by planning a path that uses cut-in and following for the dynamic object.

10. A high-definition (HD) map-based local path planning apparatus for dynamic and static object avoidance, the apparatus comprising:

a position verifier configured to verify a position on a global path by parameterizing the global path using an arc length;

a path candidate group generator configured to generate a path candidate group based on the global path; and

an optimal path selector configured to select an optimal path having a smallest cost function related to a global path following from the path candidate group,

wherein the cost function includes a global-path-following cost for selecting a path candidate group having a smallest lateral offset from the global path,

wherein the path candidate group generator is configured to determine, as an obstacle, an object that is not mapped to the HD map among objects determined through a sensor, to compute a difference in the arc length from a current vehicle to each obstacle, and to determine an arc length value of a path depending on presence or absence of the obstacle.

11. The apparatus of claim 10 , further comprising:

a vehicle velocity determiner configured to determine a velocity of a vehicle for the selected optimal path.

12. The apparatus of claim 10 , wherein the position verifier is configured to express each waypoint corresponding to the global path as coordinates by parameterizing a position of a vehicle in a road environment with respect to the arc length.

13. The apparatus of claim 10 , wherein the position verifier is configured to verify the position on the global path by expressing a position of a vehicle as an arc length and a lateral offset with respect to the global path.

14. The apparatus of claim 10 , wherein the path candidate group generator is configured to generate the path candidate group by applying a lateral offset based on the global path.

15. The apparatus of claim 10 , wherein the cost function includes a static object cost, a smoothness cost, a global-path-following cost, and a dynamic object cost, and is determined by adjusting an importance of each cost.

16. The apparatus of claim 10 , wherein the optimal path selector is configured to allow driving by reflecting a cost function for a dynamic object and by planning a path that uses cut-in and following for the dynamic object.

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (10)