Granted Patent B2

US 12,298,141 · App. 18/135,573 · Granted May 13, 2025

Inventor: Gyumin Oh (Seoul, KR)

Assignee: THORDRIVE CO., LTD

G01C21/32G01C21/005G05D1/0214G05D1/0251G05D1/0274G06V20/58

Abstract

Claims (40)

1. An autonomous driving apparatus, the apparatus comprises:

a detection sensor;

a memory; and

one or more processors in communication with the memory,

wherein the one or more processors are configured to:

extract a first statistical map from points of map data;

obtain points from the detection sensor;

extract a second statistical map from the points obtained by the detection sensor;

divide the second statistical map into a vertical-object part and a horizontal-object part depending on statistical values of respective divided regions of the second statistical map;

estimate a pose of the autonomous driving apparatus by comparing the horizontal-object part or the vertical-object part with the first statistical map; and

perform autonomous driving based on the estimated pose,

wherein regions of the statistical maps that have statistical values larger than a predetermined value are extracted as the vertical-object part, and regions of the statistical maps that have statistical values smaller than the predetermined value are extracted as the horizontal-object part.

2. The apparatus of claim 1 , comprising the detection sensor.

3. The apparatus of claim 1 , comprising the autonomous driving apparatus.

4. The apparatus of claim 1 , wherein the statistical values of respective divided regions of the second statistical map comprises at least one of a mean, a mode, a maximum, a median, a minimum, a range, an interquartile range, a quartile deviation, a variance, a standard deviation, a coefficient of variation, and a covariance.

5. The apparatus of claim 1 , in the first and second statistical maps, a 3D space is divided in accordance with a predetermined rule, and one or more statistical values of 3D points contained in a divided region represent the 3D points in the divided region.

6. The apparatus of claim 5 , wherein in the first and second statistical maps, the 3D space is divided with respect to a 2D plane and one or more statistical values of the remaining dimension of 3D points contained in the divided region are calculated and are represented on the 2D plane.

7. The apparatus of claim 6 , wherein, in the first and second statistical maps, an (x, y) plane is divided into evenly spaced grids in the Cartesian coordinate system, and one or more statistical values of height z values of 3D points corresponding to the respective grids are represented on the (x, y) plane.

8. The apparatus of claim 6 , wherein, in the first and second statistical maps, statistical values of height z values corresponding to regions formed by dividing r and θ of an (r, θ) plane evenly in the cylindrical coordinate system are represented on the (r, θ) plane.

9. The apparatus of claim 6 , wherein, in the first and second statistical maps, statistical values of respective θ values of 3D points corresponding to regions formed by dividing r and ϕ of an (r, ϕ) plane evenly in the spherical coordinate system are represented on the (r, ϕ) plane.

10. The apparatus of claim 5 , wherein the one or more processors are configured to detect whether 3D points contained in each divided region are separated into two or more groups in the statistical maps.

11. The apparatus of claim 10 , wherein the one or more processors are configured to, responsive to detecting the two or more separated groups of 3D points in a divided region, device the divided region again such that each one of divided regions contains one group of 3D points only.

12. The apparatus of claim 1 , wherein the one or more processors are configured to perform the map matching based on a location of the autonomous driving apparatus which is estimated using data obtained by a motion sensor.

13. The apparatus of claim 1 , wherein in the estimating a pose of the autonomous driving apparatus, the first statistical map is compared with the vertical-object part or the horizontal-object part, based on each points located in a search range, and the position having a highest mutual similarity is estimated as a result, wherein the position having the highest mutual similarity refers to having the lowest error between the first statistical map and the vertical-object part or the horizontal-object part.

14. The apparatus of claim 1 ,

wherein in the estimating a pose of the autonomous driving apparatus, a map matching is performed using the first statistical map and the vertical-object part, and by using the result thereof, a final map matching is performed using the first statistical map and the horizontal-object part, or

wherein in the estimating a pose of the autonomous driving apparatus, a map matching is performed using the first statistical map and the horizontal-object part, and by using the result thereof, a final map matching is performed using the first statistical map and the vertical-object part.

15. The apparatus of claim 14 ,

wherein the map matching using the vertical-object part is performed with respect to (x, y, h) or a part thereof, and the map matching using the horizontal-object part is performed with respect to (z, h, r, p) or a part thereof, or

wherein in the estimating a pose of the autonomous driving apparatus, a map matching with respect to (h, r, p) is performed using the first statistical map and the horizontal-object part to store values of (r′, p′) which minimize errors with respect to the respective h values, and a map matching with respect to (x, y, h) is performed using the first statistical map and the vertical-object part to perform a map matching with respect to (x, y, h, r′, p′) using the values of (r′, p′) with respect to the respective h values stored in advance.

16. The apparatus of claim 1 , wherein in the estimating a pose of the autonomous driving apparatus, after a map matching performed using the first statistical map and the horizontal-object part and a map matching performed using the first statistical map and the vertical-object part, the two map matching results are combined to obtain a final map matching result.

17. The apparatus of claim 16 , wherein in the estimating a pose of the autonomous driving apparatus, a map matching with respect to (x, y, h) is performed using the first statistical map and the vertical-object part to calculate the value of (x′, y′, h′) which minimize errors, and a map matching of the first statistical map and the horizontal-object part with respect to (r, p, h) is performed to store values of (r, p) which minimize errors with respect to the respective h values, and (x′, y′, h′, r′, p′) is estimated as a final position by using (r′, p′) corresponding to h′.

18. The apparatus of claim 1 , wherein the one or more processors are further configured to:

acquire coordinates of surrounding objects around the autonomous driving apparatus;

identify a temporary object by comparing positions corresponding to the coordinates of the surrounding objects in the map data in order to identify a part corresponding to the temporary object; and

remove a part corresponding to the temporary object.

19. The apparatus of claim 18 , wherein the acquiring of the coordinates of the surrounding objects around the autonomous driving apparatus further comprises:

grouping the points indicating the same object of the points input from the detection sensor, and

acquiring the coordinate of the region that the same object of the grouped points occupies.

20. The apparatus of claim 18 , wherein in the identifying of the temporary object by comparing the positions corresponding to the coordinates of the surrounding objects in the map data, a spatial feature of the corresponding coordinates is found using statistical values corresponding to the coordinates of the surrounding objects in the first statistical map, and presence of a temporary object at the corresponding coordinates is identified based on the spatial feature.

Assignments (2)

Priority Claims (1)

Continuity (2)

References Cited (28)