Object tracking in blind-spot

1. A system, comprising a processor and a memory, the memory storing instructions executable by the processor to:

identify an object, including points on the object based on image data received from vehicle sensors;

perform tracking of the object based on a motion model including a relationship of the points on the object to one another;

determine a location of one of the points in a blind spot based on the relationship of the one of the points in the blind spot to at least two points of the points on the object within a field of view of one or two vehicle sensors, wherein the blind spot is a space outside fields of view of the vehicle sensors; and

then output the determined location of the one of the points in the blind spot.

2. The system of claim 1 , wherein the motion model further includes a dead reckoning model, based on vehicle motion data, including a second relationship of the points to a vehicle reference point.

3. The system of claim 2 , wherein instructions further include instructions to track the points of the object based on at least one of dead reckoning and vehicle object detection sensor data.

4. The system of claim 1 , wherein fields of view of one or more vehicle sensors at least partially overlap.

5. The system of claim 1 , wherein instructions further include instructions to identify three-dimensional location coordinates of each of the points relative to a reference point in a vehicle.

6. The system of claim 1 , wherein the object is a trailer including a hitch attachment, and the instructions further include instructions to navigate a vehicle by actuating one or more vehicle actuators to move a vehicle hitch to a location of the hitch attachment of the trailer.

7. The system of claim 1 , wherein instructions further include instructions to track the points further based on physical attributes of a vehicle including a speed, heading, and steering angle.

8. A method, comprising:

identifying an object, including points on the object based on image data received from vehicle sensors;

performing tracking of the object based on a motion model including a relationship of the points on the object to one another;

determining a location of one of the points in a blind spot based on the relationship of the one of the points in the blind spot to at least two points of the points on the object within a field of view of one or two vehicle sensors, wherein the blind spot is a space outside fields of view of the vehicle sensors; and

then outputting the determined location of the one of the points in the blind spot.

9. The method of claim 8 , wherein the motion model further includes a dead reckoning model, based on vehicle motion data, including a second relationship of the points of the object relative to a vehicle reference point.

10. The method of claim 9 , further comprising tracking the points of the object based on at least one of dead reckoning and vehicle sensor data.

11. The method of claim 8 , wherein fields of view of one or more vehicle sensors at least partially overlap.

12. The method of claim 8 , wherein tracking the points further includes identifying three-dimensional location coordinates of each of the points relative to a reference point in a vehicle.

13. The method of claim 8 , wherein the object is a trailer including a hitch attachment, the method further comprising navigating a vehicle by actuating one or more vehicle actuators to move a vehicle hitch to a location of the hitch attachment of the trailer.

14. A system, comprising:

means for identifying an object, including points outside a vehicle blind spot, from vehicle sensor data;

means for performing tracking the object based on a motion model including a relationship of the points on the object to one another;

means for determining the location of the one of the points in the blind spot based on the relationship of the one of the points in the blind spot to at least two points of the points on the object within a field of view of one or two object detection sensors; and

means for then outputting a location of one of the points in the blind spot based on the tracking.

15. The system of claim 14 , wherein the motion model further includes a dead reckoning model, based on vehicle motion data, including a second relationship of the points of the object relative to a vehicle reference point.