Pavement marking map change detection, reaction, and live tile shipping

1. A computer-implemented method comprising:

detecting, by a light detection and ranging sensor (LiDAR) installed on an autonomous vehicle, a pavement marking around the autonomous vehicle to produce a detected pavement marking defined by points of a point cloud generated by the LiDAR;

comparing, by at least one processor, the points of the point cloud defining the detected pavement marking with corresponding data map points of a pavement marking present in a semantic data map;

determining, by the at least one processor and based on the comparing, that a change has occurred between the detected pavement marking and the pavement marking present in the semantic data map; and

updating, by the at least one processor, the semantic data map based on the change to produce an updated semantic data map, wherein the updating comprises:

generating, by the at least one processor, geospatial tiles around the autonomous vehicle, wherein the geospatial tiles are local LiDAR intensity tiles or colorized tiles;

determining a category of the detected pavement marking;

based on the category, selecting between generating the local LiDAR intensity tiles or generating the colorized tiles for the geospatial tiles, wherein the geospatial tiles include different colors based on a degree of the change between each set of the points of the point cloud corresponding to a set of the corresponding data map points, each set of the points of the point cloud corresponding to a geospatial tile of the geospatial tiles; and

designating, by the at least one processor, on the semantic data map, each geospatial tile of the geospatial tiles that is associated with the detected pavement marking.

2. The computer-implemented method of claim 1 , wherein the pavement marking is a lane line or a crosswalk.

3. The computer-implemented method of claim 1 , wherein the change that has occurred between the detected pavement marking and the pavement marking present in the semantic data map is a threshold that is exceeded between the detected pavement marking and the pavement marking present in the semantic data map.

4. The computer-implemented method of claim 1 , further comprising providing the updated semantic data map to an autonomous vehicle fleet.

5. The method of claim 1 , further comprising:

requesting, depending on the category of the detected pavement marking and the degree of the change between each set of the points of the point cloud corresponding to a set of the corresponding data map points, a level of assistance from a remote operator necessary to operate the autonomous vehicle.

6. A system comprising:

one or more processors; and

at least one computer-readable storage medium having stored therein instructions that, when executed by the one or more processors, cause the system to:

detect, by a light detection and ranging sensor (LiDAR) installed on an autonomous vehicle, a pavement marking around the autonomous vehicle to produce a detected pavement marking defined by points of a point cloud generated by the LiDAR;

compare the points of the point cloud defining the detected pavement marking with corresponding data map points of a pavement marking present in a semantic data map;

determine, based on the comparing, whether a change has occurred between the detected pavement marking and the pavement marking present in the semantic data map; and

update the semantic data map based on the determination of whether the change has occurred between the detected pavement marking and the pavement marking present in the semantic data map, wherein to update the semantic data map, the instructions, when executed by the one or more processors, further cause the system to:

generate geospatial tiles around the autonomous vehicle, wherein the geospatial tiles are local LiDAR intensity tiles or colorized tiles;

determine a category of the detected pavement marking;

based on the category, select between generating the local LiDAR intensity tiles or generating the colorized tiles for the geospatial tiles, wherein the geospatial tiles vary based on a category of the detected pavement marking and a degree of the change between each set of the points of the point cloud corresponding to a set of the corresponding data map points, wherein each set of the points of the point cloud correspond to a geospatial tile of the geospatial tiles; and

designate on the semantic data map each geospatial tile of the geospatial tiles that is associated with the detected pavement marking.

7. The system of claim 6 , wherein the pavement marking is a lane line or a crosswalk.

8. The system of claim 6 , wherein the change that has occurred between the detected pavement marking and the pavement marking present in the semantic data map is a threshold that is exceeded between the detected pavement marking and the pavement marking present in the semantic data map.

9. The system of claim 6 , wherein the instructions, when executed by the one or more processors, further cause the system to provide the updated semantic data map to an autonomous vehicle fleet.

10. The system of claim 6 , wherein the category of the detected pavement marking is determined by at least one of a type of the pavement marking, a purpose for the pavement marking, or a threshold level of change to the pavement marking.

11. The system of claim 6 , wherein the instructions, when executed by the one or more processors, further cause the system to:

request, depending on the category of the detected pavement marking and the degree of the change between each set of the points of the point cloud corresponding to a set of the corresponding data map points, a level of assistance from a remote operator necessary to operate the autonomous vehicle.

12. A non-transitory computer-readable storage medium comprising:

instructions, stored on the non-transitory computer-readable storage medium, that, when executed by one or more processors, cause the one or more processors to:

detect, by a light detection and ranging sensor (LiDAR) installed on an autonomous vehicle, a pavement marking around the autonomous vehicle to produce a detected pavement marking defined by points of a point cloud generated by the LiDAR;

compare the points of the point cloud defining the detected pavement marking with corresponding data map points of a pavement marking present in a semantic data map;

determine, based on the comparing, whether a change has occurred between the detected pavement marking and the pavement marking present in the semantic data map; and

update the semantic data map based on the determination of whether the change has occurred between the detected pavement marking and the pavement marking present in the semantic data map, wherein to update the semantic data map, the instructions, when executed by the one or more processors, further cause the one or more processors to:

generate geospatial tiles around the autonomous vehicle, wherein the geospatial tiles are local LiDAR intensity tiles or colorized tiles;

determine a category of the detected pavement marking;

based on the category, select between generating the local LiDAR intensity tiles or generating the colorized tiles for the geospatial tiles, wherein the geospatial tiles vary based on a category of the detected pavement marking and a degree of the change between each set of the points of the point cloud corresponding to a set of the corresponding data map points, wherein each set of the points of the point cloud correspond to a geospatial tile of the geospatial tiles; and

designate on the semantic data map each geospatial tile of the geospatial tiles that is associated with the detected pavement marking.

13. The non-transitory computer-readable storage medium of claim 12 , wherein the pavement marking is a lane line or a crosswalk.

14. The non-transitory computer-readable storage medium of claim 12 , wherein the change that has occurred between the detected pavement marking and the pavement marking present in the semantic data map is a threshold that is exceeded between the detected pavement marking and the pavement marking present in the semantic data map.

15. The non-transitory computer-readable storage medium of claim 12 , wherein the instructions, when executed by the one or more processors, further cause the one or more processors to provide the updated semantic data map to an autonomous vehicle fleet.

16. The non-transitory computer-readable storage medium of claim 15 , wherein the category of the detected pavement marking is determined by at least one of a type of the pavement marking, a purpose for the pavement marking, or a threshold level of change to the pavement marking.

17. The non-transitory computer readable storage medium of claim 12 , wherein the instructions, when executed by the one or more processors, further cause the one or more processors to:

request, depending on the category of the detected pavement marking and the degree of the change between each set of the points of the point cloud corresponding to a set of the corresponding data map points, a level of assistance from a remote operator necessary to operate the autonomous vehicle.