US 12,252,150 B2
Method and apparatus for tracking object using LiDAR sensor and recording medium storing program to execute the method
Nam Gyun Kim, Seongnam-si (KR); and Mu Gwan Jeong, Seoul (KR)
Assigned to Hyundai Motor Company, Seoul (KR); and Kia Corporation, Seoul (KR)
Filed by Hyundai Motor Company, Seoul (KR); and Kia Corporation, Seoul (KR)
Filed on Mar. 22, 2022, as Appl. No. 17/701,290.
Claims priority of application No. 10-2021-0057744 (KR), filed on May 4, 2021.
Prior Publication US 2022/0363288 A1, Nov. 17, 2022
Int. Cl. B60W 60/00 (2020.01); G06V 20/58 (2022.01)
CPC B60W 60/001 (2020.02) [G06V 20/58 (2022.01); B60W 2420/408 (2024.01); B60W 2554/4041 (2020.02); B60W 2554/4049 (2020.02)] 17 Claims
OG exemplary drawing
 
1. A method for tracking an object, the method comprising:
checking points generated by a LiDAR sensor to determine a first interpolation reference point and a second interpolation reference point among the points generated by the LiDAR when points, associated with the object, are missing,
wherein the first interpolation reference point is determined by:
determining a candidate for the first interpolation reference point,
checking whether the candidate is a point that is recognizable as the object, and
determining that the candidate is the first interpolation reference point based on a result of the checking, and
wherein the second interpolation reference point is determined by:
subsequent to determining the first interpolation reference point, checking whether there is a missing point, and
when the missing point exists, generating another point and determining that the other point is the second interpolation reference point;
generating interpolation points to replace the missing points between the first interpolation reference point and the second interpolation reference point;
performing primary labeling on segmented points including the interpolation points to generate first clusters;
checking a density of the interpolation points to find a target cluster on which secondary labeling is to be performed among the first clusters; and
performing the secondary labeling on the target cluster to generate a second cluster, wherein the secondary labeling is performed on the target cluster after excluding the interpolation points.