Detection of a change in a drivable area

1 . A method, comprising:

determining, by one or more computing devices, a ground height associated with a cell of a tile map based on sensor data from a sensor of a vehicle;

comparing, by the one or more computing devices, the ground height with data stored in a base map defining a drivable area;

updating, by the one or more computing devices, a dynamic vector map in response to a first determination that a difference in the ground height is greater than a first threshold, wherein the difference in the ground height is determined using the data stored in the base map, and the dynamic vector map includes changes to the base map detected during an operation of the vehicle; and

updating, by the one or more computing devices, the base map based on a second determination that there is a change in the drivable area of the vehicle and the change is permanent, the second determination being based on at least the ground height and the data stored in the base map.

2 . The method of claim 1 , further comprising:

assigning, by the one or more computing devices, an attribute indicating the change in the drivable area to the cell when the difference in the ground height is greater than the first threshold; and

identifying, by the one or more computing devices, a number of cells of the tile map that are associated with the attribute indicating the change; and

wherein the dynamic vector map is updated when the number of cells is greater than a second threshold.

3 . The method of claim 1 , further comprising:

clustering, by the one or more computing devices, cells into a cluster of the tile map that indicate a new structure in the drivable area or that a structure is removed; and

calculating, by the one or more computing devices, a polygon for the cluster indicating an area of the new structure or the area indicating where the structure have been removed.

4 . The method of claim 1 , wherein the sensor data includes a plurality of LiDAR points, the method further comprising:

filtering, by the one or more computing devices, a LiDAR point of the plurality of LiDAR points in response to a determination that the LiDAR point is associated with a dynamic object.

5 . The method of claim 4 , wherein the sensor data includes a semantic label associated with the LiDAR point of the plurality of LiDAR points, the method further comprising:

filtering, by the one or more computing devices, the LiDAR point based on the semantic label.

6 . A system, comprising:

a memory; and

at least one processor coupled to the memory and configured to perform operations comprising:

determining a ground height associated with a cell of a tile map based on sensor data from a sensor of a vehicle,

comparing the ground height with data stored in a base map defining a drivable area,

updating a dynamic vector map in response to a first determination that a difference in the ground height is greater than a first threshold, wherein the difference in the ground height is determined using the data stored in the base map, and the dynamic vector map includes changes to the base map detected during an operation of the vehicle; and

updating the base map in response to a second determination that there is a change in the drivable area of the vehicle and the change is permanent, the second determination being based on at least the ground height and the data stored in the base map.

7 . The system of claim 6 , the operations further comprising:

assigning an attribute indicating the change in the drivable area to the cell when the difference in the ground height is greater than the first threshold;

identifying a number of cells of the tile map that are associated with the attribute indicating the change; and

a wherein the dynamic vector map is updated when the number of cells is greater than a second threshold.

8 . The system of claim 6 , the operations further comprising:

clustering cells of the tile map into a cluster that indicate a new structure in the drivable area; and

calculating a polygon for the cluster indicating an area of the new structure.

9 . The system of claim 6 , wherein the sensor data includes a plurality of LiDAR points,

the operations further comprising:

filtering a LiDAR point of the plurality of LiDAR points in response to a determination that the LiDAR point is associated with a dynamic object.

10 . The system of claim 9 , wherein the sensor data includes a semantic label associated with the LiDAR point of the plurality of LiDAR points, the operations further comprising:

filtering the LiDAR point based on the semantic label.

11 . A non-transitory computer-readable medium having instructions stored thereon that, when executed by at least one computing device, cause the at least one computing device to perform operations comprising:

determining a ground height associated with a cell of a tile map based on sensor data from a sensor of a vehicle;

comparing the ground height with data stored in a base map defining a drivable area;

updating a dynamic vector map in response to a first determination that a difference in the ground height is greater than a first threshold, wherein the difference in the ground height is determined using the data stored in the base map, and the dynamic vector map includes changes to the base map detected during an operation of the vehicle; and

updating the base map based on a second determination that there is a change in the drivable area and the change is permanent, the second determination being based on at least the ground height and the data stored in the base map.

12 . The non-transitory computer-readable medium of claim 11 , the operations further comprising:

assigning an attribute indicating the change in the drivable area to the cell when the difference in the ground height is greater than the first threshold;

identifying a number of cells of the tile map that are associated with the attribute indicating the change; and

a wherein the dynamic map vector is updated when the number of cells is greater than a second threshold.

13 . The non-transitory computer-readable medium of claim 11 , the operations further comprising:

clustering cells of the tile map into a cluster that indicate a new structure in the drivable area; and

calculating a polygon for the cluster indicating an area of the new structure.

14 . The non-transitory computer-readable medium of claim 11 , wherein the sensor data includes a plurality of LiDAR points and wherein the operations further comprising:

filtering a LiDAR point of the plurality of LiDAR points in response to a determination that the LiDAR point is associated with a dynamic object.