Ground classifier system for automated vehicles

1. A ground-classifier system that classifies a ground-cover proximate to an automated vehicle, said system comprising:

a lidar that detects a point-cloud of a field-of-view about the automated vehicle;

a camera that renders an image of the field-of-view; and

a controller in communication with the camera and the lidar, said controller configured to:

define a lidar-grid that segregates the point-cloud into an array of patches;

define a camera-grid that segregates the image into an array of cells, wherein the point-cloud and the image are aligned such that a patch in the array of patches is aligned with a cell in the array of cells;

determine height values for cloud-points within the patch, wherein the patch is determined to be ground-cover when the height values are less than a height-threshold;

determine intensity values for the cloud-points within the patch;

determine temperature values for pixels within the cell; and

determine, when the patch is determined to be ground-cover, a ground-cover classification of the patch based on the intensity values of the cloud points within the patch and the temperature values of the pixels within the cell.

2. The system in accordance with claim 1 , wherein the classification of the patch is determined to be asphalt when:

the intensity values of the cloud points within the patch are less than an intensity-threshold; and

saturation values of the pixels within the cell indicated by the camera are less than a saturation-threshold.

3. The system in accordance with claim 1 , wherein the classification of the patch is determined to be grass when:

the intensity values of the cloud points within the patch are greater than an intensity-threshold; and

hue values of the pixel within the cell indicated by the camera are within a hue-range.

4. The system in accordance with claim 1 , wherein the classification of the patch is determined to be lane-marking when:

the intensity values of the cloud points within the patch are greater than an intensity-threshold; and

brightness values of the pixels within the cell indicated by the camera are greater than a brightness-threshold.

5. The system in accordance with claim 1 , wherein the classification of the patch is determined to be snow when:

the intensity values of the cloud points within the patch are greater than an intensity-threshold; and

saturation values of the pixels within the cell indicated by the camera are less than a saturation-threshold.

6. The system in accordance with claim 1 , wherein:

the controller is further configured to determine intensity values and temperature values for cloud-points of an adjacent patch and pixels of an adjacent cell, respectively; and

the classification of the patch is further based on the intensity values and the temperature values of the adjacent patch and the adjacent cell, respectively.

7. The system in accordance with claim 6 , wherein the controller is further configured to determine a classification of the adjacent patch.

8. The system in accordance with claim 7 , wherein the controller is further configured to form a cluster when the classification of the patch and the classification of the adjacent patch are equivalent.

9. The system in accordance with claim 6 , wherein the classification of the patch is determined to be asphalt when the temperature values for the cell are generally more than a threshold amount greater than the temperature values for the adjacent cell.

10. The system in accordance with claim 6 , wherein the classification of the patch is determined to be grass when the temperature values for the cell are generally more than a threshold amount less than the temperature values for the adjacent cell.

11. A method of patch classification performed by a vehicle, the method comprising:

defining a lidar grid that segregates point cloud data from a lidar system of the vehicle into an array of patches;

defining a camera grid that segregates image data from a camera system of the vehicle into an array of cells, the camera grid aligned with the lidar grid such that the cells align with the patches;

determining azimuth angles and elevation angles for cloud-points within a patch of the patches based on the point cloud data corresponding to the patch;

determining height values for the cloud-points within the patch based on the azimuth angles and the elevation angles; and

responsive to the height values for the cloud-points within the patch being less than a height threshold:

classifying the patch as ground cover;

determining intensity values for the cloud-points within the patch;

determining temperature values for pixels within a cell of the array of cells that corresponds to the patch based on the image data corresponding to the cell; and

determining a ground cover classification of the patch based on the intensity values of the cloud points within the patch and the temperature values of the pixels within the cell.

12. The method of claim 11 , wherein the ground cover classification of the patch is determined to be asphalt when:

the intensity values for the cloud points within the patch are generally less than an intensity threshold; and

saturation values for the pixels within the cell indicated by the image data are generally less than a saturation threshold.

13. The method of claim 11 , wherein the ground cover classification of the patch is determined to be grass when:

the intensity values for the cloud points within the patch are generally greater than an intensity threshold; and

hue values for the pixels within the cell indicated by the image data are generally within a hue-range.

14. The method of claim 11 , wherein the ground cover classification of the patch is determined to be lane marking when:

the intensity values for the cloud points within the patch are generally greater than an intensity threshold; and

brightness values for the pixels within the cell indicated by the image data are generally greater than a brightness-threshold.

15. The method of claim 11 , wherein the ground cover classification of the patch is determined to be snow when:

the intensity values for the cloud points within the patch are generally greater than an intensity threshold; and

saturation values for the pixels within the cell indicated by the image data are generally less than a saturation threshold.

16. The method of claim 11 :

further comprising:

determining adjacent intensity values for cloud points of an adjacent patch; and

determining adjacent temperature values for pixels of an adjacent cell that corresponds to the patch; and

wherein the ground-cover classification of the patch is further based on the adjacent intensity values and adjacent temperature values.

17. The method of claim 16 , further comprising determining an adjacent classification for the adjacent patch.

18. The method of claim 17 , further comprising forming a cluster when the classification of the patch and the adjacent classification of the adjacent-patch are equivalent.

19. The method of claim 16 , wherein the classification of the patch is determined to be asphalt when the temperature values for the pixels of the cell are generally more than a threshold amount greater than the adjacent temperature values for the pixels of the adjacent cell.

20. The method of claim 16 , wherein the classification of the patch is determined to be grass when the temperature values for the pixels of the cell are generally more than a threshold amount less than the adjacent temperature values for the pixels of the adjacent cell.