USING PREDICTIVE VISUAL ANCHORS TO CONTROL AN AUTONOMOUS VEHICLE

1 . A method comprising:

determining, based on a plurality of frames of video data from a camera of an autonomous vehicle, one or more predicted visual anchors, wherein the one or more predicted visual anchors comprise a predicted location of one or more visual anchors at a future time relative to when the plurality of frames were captured;

identifying, in another frame of video data corresponding to the future time, the one or more visual anchors;

determining one or more differentials between the one or more visual anchors and the one or more predicted visual anchors;

determining, based on the one or more differentials, one or more control operations for the autonomous vehicle; and

applying the one or more control operations.

2 . The method of claim 1 , wherein determining the one or more control operations comprises determining the one or more control operations to reduce the one or more differentials.

3 . The method of claim 1 , wherein the one or more differentials comprise one or more pixel width differentials, one or more pixel height differentials, one or more pixel area differentials, or one or more vector differentials.

4 . The method of claim 1 , wherein determining the one or more control operations is based on a rate of change of the one or more differentials.

5 . The method of claim 1 , wherein determining the one or more control operations comprises providing the one or more differentials to a controller of a control system of the autonomous vehicle.

6 . The method of claim 5 , wherein the controller interfaces with one or more actuators of the autonomous vehicle.

7 . The method of claim 1 , wherein determining the one or more predicted visual anchors comprises providing the plurality of frames of video data as input to one or more machine learning models.

8 . An apparatus comprising a computer processor, a computer memory operatively coupled to the computer processor, the computer memory having disposed within it computer program instructions that, when executed by the computer processor, cause the apparatus to carry out steps comprising:

determining, based on a plurality of frames of video data from a camera of an autonomous vehicle, one or more predicted visual anchors, wherein the one or more predicted visual anchors comprise a predicted location of one or more visual anchors at a future time relative to when the plurality of frames were captured;

identifying, in another frame of video data corresponding to the future time, the one or more visual anchors;

determining one or more differentials between the one or more visual anchors and the one or more predicted visual anchors;

determining, based on the one or more differentials, one or more control operations for the autonomous vehicle; and

applying the one or more control operations.

9 . The apparatus of claim 8 , wherein determining the one or more control operations comprises determining the one or more control operations to reduce the one or more differentials.

10 . The apparatus of claim 8 , wherein the one or more differentials comprise one or more pixel width differentials, one or more pixel height differentials, one or more pixel area differentials, or one or more vector differentials.

11 . The apparatus of claim 8 , wherein determining the one or more control operations is based on a rate of change of the one or more differentials.

12 . The apparatus of claim 8 , wherein determining the one or more control operations comprises providing the one or more differentials to a controller of a control system of the autonomous vehicle.

13 . The apparatus of claim 12 , wherein the controller interfaces with one or more actuators of the autonomous vehicle.

14 . The apparatus of claim 8 , wherein determining the one or more predicted visual anchors comprises providing the plurality of frames of video data as input to one or more machine learning models.

15 . A computer program product disposed upon a non-transitory computer readable medium, the computer program product comprising computer program instructions that, when executed, cause a computer system of the autonomous vehicle to perform a method comprising:

determining, based on a plurality of frames of video data from a camera of an autonomous vehicle, one or more predicted visual anchors, wherein the one or more predicted visual anchors comprise a predicted location of one or more visual anchors at a future time relative to when the plurality of frames were captured;

identifying, in another frame of video data corresponding to the future time, the one or more visual anchors;

determining one or more differentials between the one or more visual anchors and the one or more predicted visual anchors;

determining, based on the one or more differentials, one or more control operations for the autonomous vehicle; and

applying the one or more control operations.

16 . The computer program product of claim 15 , wherein determining the one or more control operations comprises determining the one or more control operations to reduce the one or more differentials.

17 . The computer program product of claim 15 , wherein the one or more differentials comprise one or more pixel width differentials, one or more pixel height differentials, one or more pixel area differentials, or one or more vector differentials.

18 . The computer program product of claim 15 , wherein determining the one or more control operations is based on a rate of change of the one or more differentials.

19 . The computer program product of claim 15 , wherein determining the one or more control operations comprises providing the one or more differentials to a controller of a control system of the autonomous vehicle.

20 . The computer program product of claim 19 , wherein the controller interfaces with one or more actuators of the autonomous vehicle.