Method of adjusting grid spacing of height map for autonomous driving

1. A method of adjusting a grid spacing of a height map for autonomous driving, the method comprising:

acquiring a 2D image of a region ahead of a vehicle;

generating a depth map using depth information on an object present in the 2D image and converting the generated depth map into a 3D point cloud;

generating the height map by mapping the 3D point cloud onto a grid having a predetermined size; and

adjusting the grid spacing of the height map according to a driving state of the vehicle relative to the object,

wherein the generating the height map includes determining a mean height using a number of 3D point clouds mapped onto each grid of the height map and height information,

wherein the method further includes determining whether the object is a dynamic obstacle by accumulating a previous frame on a current frame of the height map according to motion information of the vehicle, and

wherein the determining whether the object is a dynamic obstacle includes:

performing random variable standardization in association with a mean height of each of cells forming the grid; and

comparing a standardized random variable with a boundary value of a specific standard deviation to detect a cell in which the dynamic obstacle is located, among the cells,

wherein the dynamic obstacle is in plural to include a plurality of dynamic obstacles, and

wherein the adjusting the grid spacing of the height map includes:

determining a priority of each of the dynamic obstacles, in consideration of each movement direction of the dynamic obstacles; and

adjusting differently grid spacing of each of the cells according to the determined priority.

2. The method of claim 1 , further including:

determining at least one of a relative distance between the dynamic obstacle and the vehicle or a time to collision of the vehicle with the dynamic obstacle.

3. The method of claim 2 , wherein the adjusting the grid spacing of the height map includes:

when the relative distance is less than or equal to a predetermined first threshold value, adjusting a grid spacing of a cell in which the dynamic obstacle is located and a grid spacing of a neighboring cell adjacent to the cell in a first grid adjustment stage, among the cells.

4. The method of claim 3 , wherein the adjusting the grid spacing of the height map further includes:

when the time to collision is less than or equal to a predetermined second threshold value, adjusting a grid spacing of a cell in which the dynamic obstacle is located and a grid spacing of a neighboring cell adjacent to the cell in a second grid adjustment stage, among the cells.

5. The method of claim 4 , wherein a grid spacing in the first grid adjustment stage and a grid spacing in the second grid adjustment stage are different from each other.

6. A non-transitory computer-readable recording medium in which a program configured to be executed by a processor to perform the method of claim 1 is recorded.

7. An apparatus of adjusting a grid spacing of a vehicular map, the apparatus comprising:

a sensor configured to acquire a 2D image of a region ahead of a vehicle;

an image processor configured to generate a depth map using depth information on an object present in the 2D image, to convert the generated depth map into a 3D point cloud, and to generate a height map by mapping the 3D point cloud onto a grid having a predetermined size; and

a grid spacing adjuster including a processor configured to adjust a grid spacing of the height map according to a driving state of the vehicle relative to the object,

wherein the image processor is configured to determine a mean height using a number of 3D point clouds mapped onto each grid of the height map and height information,

wherein the apparatus further includes a map analyzer including a processor configured to determine whether the object is a dynamic obstacle by accumulating a previous frame on a current frame of the height map according to motion information of the vehicle,

wherein the map analyzer is configured to perform random variable standardization in association with a mean height of each of cells forming the grid, and to compare a standardized random variable with a boundary value of a specific standard deviation to detect a cell in which the dynamic obstacle is located, among the cells,

wherein the dynamic obstacle is in plural to include a plurality of dynamic obstacles, and

wherein the grid spacing adjuster is configured to determine a priority of each of the dynamic obstacles, in consideration of each movement direction of the dynamic obstacles and to adjust grid spacing of each of the cells differently according to the determined priority.

8. The apparatus of claim 7 , further including:

a calculator including a processor configured to determine at least one of a relative distance between the dynamic obstacle and the vehicle or a time to collision of the vehicle with the dynamic obstacle.

9. The apparatus of claim 8 , wherein, when the relative distance is less than or equal to a predetermined first threshold value, the grid spacing adjuster is configured to adjust a grid spacing of a cell in which the dynamic obstacle is located and a grid spacing of a neighboring cell adjacent to the cell in a first grid adjustment stage, among the cells.

10. The apparatus of claim 9 , wherein, when the time to collision is less than or equal to a predetermined second threshold value, the grid spacing adjuster is configured to adjust a grid spacing of a cell in which the dynamic obstacle is located and a grid spacing of a neighboring cell adjacent to the cell in a second grid adjustment stage, among the cells.

11. The apparatus of claim 10 , wherein a grid spacing in the first grid adjustment stage and a grid spacing in the second grid adjustment stage are different from each other.