Tow vehicle and trailer alignment

1. A method of aligning a tow vehicle with a trailer positioned behind the tow vehicle, the method comprising:

determining, by data processing hardware, a point cloud map of a rear environment of the tow vehicle, the rear environment including the trailer;

determining, by the data processing hardware, a front face plane of the trailer based on the point cloud map;

determining, by the data processing hardware, a normal line of the front face plane; and

determining, by the data processing hardware, a path from the tow vehicle to the trailer such that a fore-aft axis of the tow vehicle is aligned with the normal line of the front face plane.

2. The method of claim 1 , further comprising:

receiving one or more images from a camera positioned on a back portion of the tow vehicle and in communication with the data processing hardware; and

identifying a region of interest (ROI) that includes a trailer representation within the one or more images, the trailer representation indicative of the trailer positioned behind the tow vehicle;

wherein determining the point cloud map of the rear environment of the tow vehicle includes determining the point cloud map associated with the ROI only.

3. The method of claim 1 , further comprising:

receiving images from a camera positioned on a back portion of the tow vehicle and in communication with the data processing hardware;

determining at least three three-dimensional positions of the rear camera from the received images, while the tow vehicle is moving; and

determining a camera plane based on the at least three three-dimensional positions.

4. The method of claim 3 , wherein determining the front face of the trailer based on the point cloud map includes:

determining clusters of points within the point cloud map;

for each cluster, determining a center of the cluster;

projecting the center of the cluster on the camera plane;

determining the cluster associated with the center having the shortest distance to a center of the camera; and

determining the front face plane based on the points of the cluster of points having the shortest distance to the center of the camera.

5. The method of claim 4 , further comprising:

projecting the points of the cluster associated with the center having the shortest distance to the center of the camera on the camera plane;

determining a line associated with the projected points; and

projecting the projected points on the line to determine a segment containing all the projected points, a line through a midpoint of the segment along the normal line is indicative of an alignment direction.

6. The method of claim 1 , wherein determining the point cloud map comprises executing one of a Visual Odometry (VO) algorithm, a Simultaneous Localization and Mapping (SLAM) algorithm, and a Structure from Motion (SfM) algorithm.

7. The method of claim 1 , further comprising:

sending instructions to a drive system in communication with the data processing hardware, the instructions causing the tow vehicle to autonomously drive along the path in a rearward direction.

8. A system for aligning a tow vehicle with a trailer positioned behind the tow vehicle, the system comprising:

data processing hardware; and

memory hardware in communication with the data processing hardware, the memory hardware storing instructions that when executed on the data processing hardware cause the data processing hardware to perform operations comprising:

determining a point cloud map of a rear environment of the tow vehicle, the rear environment including the trailer;

determining a front face plane of the trailer based on the point cloud map;

determining a normal line of the front face plane; and

determining a path from the tow vehicle to the trailer such that a fore-aft axis of the tow vehicle is aligned with the normal line of the front face plane.

9. The system of claim 8 , wherein the operations further include:

receiving one or more images from a camera positioned on a back portion of the tow vehicle and in communication with the data processing hardware; and

identifying a region of interest (ROI) that includes a trailer representation within the one or more images, the trailer representation indicative of the trailer positioned behind the tow vehicle;

wherein determining the point cloud map of the rear environment of the tow vehicle includes determining the point cloud map associated with the ROI only.

10. The system of claim 8 , wherein the operations further include:

receiving images from a camera positioned on a back portion of the tow vehicle and in communication with the data processing hardware;

determining at least three three-dimensional positions of the rear camera from the received images, while the tow vehicle is moving; and

determining a camera plane based on the at least three three-dimensional positions.

11. The system of claim 10 , wherein determining the front face of the trailer based on the point cloud map includes:

determining clusters of points within the point cloud map;

for each cluster, determining a center of the cluster;

projecting the center of the cluster on the camera plane;

determining the cluster associated with the center having the shortest distance to a center of the camera; and

determining the front face plane based on the points of the cluster of points having the shortest distance to the center of the camera.

12. The system of claim 11 , wherein the operations further include:

projecting the points of the cluster associated with the center having the shortest distance to the center of the camera on the camera plane;

determining a line associated with the projected points; and

projecting the projected points on the line to determine a segment containing all the projected points, a line through a midpoint of the segment along the normal line is indicative of an alignment direction.

13. The system of claim 8 , wherein determining the point cloud map comprises executing one of a Visual Odometry (VO) algorithm, a Simultaneous Localization and Mapping (SLAM) algorithm, and a Structure from Motion (SfM) algorithm.

14. The system of claim 8 , wherein the operations further include:

sending instructions to a drive system in communication with the data processing hardware, the instructions causing the tow vehicle to autonomously drive along the path in a rearward direction.