Systems and methods for prioritizing object prediction for autonomous vehicles

1. A computer-implemented method, comprising:

obtaining, by a computing system comprising one or more processors, state data descriptive of at least a current state or a past state of a plurality of objects that are perceived by an autonomous vehicle;

determining, by the computing system, a priority classification for each object of the plurality of objects based at least in part on respective state data for such object;

determining, by the computing system, an order at which the computing system determines a predicted future state for each object based at least in part on the priority classification for one or more of the plurality of objects;

determining, by the computing system, the predicted future state for each object of the plurality of objects based at least in part on the order, wherein determining the predicted future state comprises determining the predicted future state for objects having a higher priority classification prior to determining the predicted future state for objects having a lower priority classification; and

initiating, by the computing system, a motion of the autonomous vehicle based, at least in part, on the predicted future state of at least one of the plurality of objects.

2. The computer-implemented method of claim 1 , wherein:

the state data includes an object type; and

determining, by the computing system, the priority classification for each object is based at least in part on a respective object type of such object.

3. The computer-implemented method of claim 1 , wherein determining, by the computing system, the order at which the computing system determines the predicted future state for the plurality of objects, comprises:

prioritizing the plurality of objects for object prediction based on the priority classification for each object of the plurality of objects, wherein prioritizing the plurality of objects includes prioritizing objects having the higher priority classification for object prediction prior to objects having the lower priority classification.

4. The computer-implemented method of claim 1 , further comprising:

providing, by the computing system to a motion planning system, the predicted future state for each object of the plurality of objects having the higher priority classification for a current time frame; and

providing, by the computing system to a motion planning system, the predicted future state for each object of the plurality of objects having the lower priority classification for a previous time frame concurrently with the predicted future state for each object having the higher priority classification for the current time frame.

5. The computer-implemented method of claim 1 , wherein obtaining, by the computing system, state data descriptive of at least the current state or the past state of the plurality of objects that are perceived by the autonomous vehicle comprises obtaining, by the computing system for a plurality of consecutive time frames, state data descriptive of the current state of the plurality of objects that are perceived by the autonomous vehicle.

6. The computer-implemented method of claim 5 , wherein determining, by the computing system, the predicted future state for objects having the higher priority classification comprises:

determining, by the computing system, the predicted future state for objects having the higher priority classification based at least in part on state data obtained for a current time frame.

7. The computer-implemented method of claim 5 , wherein determining, by the computing system, the predicted future state for objects having the lower priority classification comprises:

determining, by the computing system, the predicted future state for objects having the lower priority classification based at least in part on state data obtained for a previous time frame.

8. The computer-implemented method of claim 1 , wherein obtaining, by the computing system, state data descriptive of at least the current state or the past state of the plurality of objects that are perceived by the autonomous vehicle comprises:

obtaining the state data from a perception system of the autonomous vehicle configured to generate the state data based on sensor data obtained from one or more sensors of the autonomous vehicle.

9. The computer-implemented method of claim 1 , further comprising:

selecting, by the computing system, a future location prediction system based at least in part on the priority classification for at least one of the plurality of objects.

10. The computer-implemented method of claim 9 , wherein:

determining, by the computing system, the predicted future state for the at least one of the plurality of objects comprises using the selected future location prediction system.

11. The computer-implemented method of claim 10 , wherein:

the future location prediction system comprises either a first prediction system or a second prediction system, the second prediction system having a higher fidelity than the first prediction system;

determining, by the computing system, the predicted future state for each object having the lower priority classification comprises determining, by the computing system, the predicted future state for such object using the first prediction system; and

determining, by the computing system, the predicted future state for each object having the higher priority classification comprises determining, by the computing system, the predicted future state for such object using the second prediction system.

12. The computer-implemented method of claim 1 , wherein determining, by the computing system, the priority classification for each object comprises:

determining a number of the plurality of objects to be classified as high-priority objects based on a threshold velocity or a velocity range of the autonomous vehicle.

13. The computer-implemented method of claim 1 , wherein the current state or the past state of each object of the plurality of objects comprises one or more of:

a position, a velocity, an acceleration, a heading, a yaw rate, a shape, a size, a type, a distance from the autonomous vehicle, a minimum path to interaction with the autonomous vehicle, or a minimum time duration to interaction with the autonomous vehicle.

14. A computing system, comprising:

a perception system comprising one or more processors, wherein the perception system is configured to generate, for each of a plurality of consecutive time frames, state data descriptive of at least a current state of each of a plurality of objects that are perceived by an autonomous vehicle;

a priority classification system comprising one or more processors, wherein the priority classification system is configured to, for each of the plurality of consecutive time frames, assign a priority classification to each object of the plurality of objects based at least in part on the respective state data for each object;

a prediction system comprising one or more processors, wherein the prediction system is configured to, for each of the plurality of consecutive time frames:

receive the priority classification for each object of the plurality of objects;

determine, for a current time frame, a predicted future state for each object of the plurality of objects;

provide the predicted future state for each object of the plurality of objects to a motion planning system implemented by the one or more processors, wherein the predicted future state for each object having a higher priority classification is provided to the motion planning system prior to the predicted future state for each object having a lower priority classification; and

a vehicle controller configured to control one or more vehicle controls associated with a motion of the autonomous vehicle based, at least in part, on the predicted future state of at least one of the plurality of objects.

15. The computing system of claim 14 , wherein the prediction system is further configured to:

provide, to the motion planning system, a predicted future state for a previous time frame for each object having the lower priority classification; and

wherein the predicted future state for the previous time frame for each object having the lower priority classification is provided to the motion planning system concurrently with the predicted future state for each object having the higher priority classification for the current time frame.

16. The computing system of claim 14 , wherein the prediction system is configured to:

determine, for the current time frame, the predicted future state for each object having the higher priority classification prior to determining the predicted future state for each object having the lower priority classification.

17. The computing system of claim 14 , wherein the prediction system is configured to:

determine, for the current time frame, the predicted future state for each object having the higher priority classification based on state data for the current time frame; and

determine, for the current time frame, the predicted future state for each object having the lower priority classification based on state data for a previous time frame.

18. The computing system of claim 14 , wherein the prediction system is further configured to:

perform a low-fidelity prediction for each object having the lower priority classification; and

perform a high-fidelity prediction for each object having the higher priority classification.

19. An autonomous vehicle, comprising:

one or more processors; and

one or more non-transitory computer-readable media that collectively store instructions that, when executed by the one or more processors, cause the one or more processors to perform operations, the operations comprising:

obtaining state data descriptive of an object type of a plurality of objects that are perceived by an autonomous vehicle;

determining a priority classification for each object of the plurality of objects based at least in part on the respective object type for such object;

determining an order at which the one or more processors determines a predicted future state for each object based at least in part on the priority classification for one or more of the plurality of objects;

determining the predicted future state for each object of the plurality of objects based at least in part on the order, wherein determining the predicted future state comprises determining the predicted future state for objects having a higher priority classification prior to determining the predicted future state for objects having a lower priority classification; and

initiating a motion of the autonomous vehicle based, at least in part, on the predicted future state of at least one of the plurality of objects.

20. The autonomous vehicle of claim 19 , wherein the operations further comprise:

providing, to a motion planning system comprising the one or more processors, the predicted future state for each object of the plurality of objects having the higher priority classification for a current time frame; and

providing the predicted future state for each object of the plurality of objects having the lower priority classification for a previous time frame concurrently with the predicted future state for each object having the higher priority classification for the current time frame.