| CPC G01S 13/89 (2013.01) [G06N 3/0455 (2023.01); B60W 60/001 (2020.02)] | 13 Claims |

|
1. A method of compressing normal distribution transform (NDT) map data generated by modeling a three-dimensional (3D) point cloud for a predetermined area as a normal distribution set for an autonomous driving system, which is performed by a computing device, the method comprising:
collecting the 3D point cloud for the predetermined area through sensors mounted on an autonomous driving vehicle travelling within the predetermined area;
generating a plurality of 3D lattice spaces by latticing the 3D point cloud for the predetermined area;
modeling each of the 3D point clouds included in each of the plurality of generated 3D lattice spaces as a plurality of normal distributions;
calculating a mean vector for each of the plurality of normal distributions corresponding to each of the plurality of generated 3D lattice spaces, and generating mean vector data using the calculated mean vector;
calculating a covariance matrix for each of the plurality of normal distributions corresponding to each of the generated 3D lattice spaces, and generating covariance matrix data using the calculated covariance matrix;
generating NDT map data for the predetermined area using the generated mean vector data and the generated covariance matrix data,
wherein the generated mean vector data includes a center vector for a center position of each of the plurality of generated 3D lattice spaces and an offset vector from the center position;
processing the generated mean vector data included in the NDT map data;
processing the generated covariance matrix data included in the NDT map data;
generating compressed NDT map data using the processed mean vector data and the processed covariance matrix data; and
controlling driving operations of the autonomous driving vehicle by using the compressed NDT map data,
wherein the processing of the generated mean vector data includes:
transforming an expression method of the center vector;
transforming an expression method of the offset vector; and
generating compressed mean vector data using the transformed center vector and the transformed offset vector.
|