Systems and methods for streaming processing for autonomous vehicles

1. An autonomy computing system comprising:

one or more processors; and

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

(a) accessing autonomy data descriptive of a state of a first object and a state of a second object in a surrounding environment of an autonomous vehicle;

(b) determining an interaction classification associated with the first object relative to the second object, the interaction classification indicative of a likelihood that the first object and the second object will interact;

(c) obtaining a first portion of additional autonomy data descriptive of an updated state of the first object;

(d) determining an expected time estimate for receiving a second portion of additional autonomy data descriptive of an updated state of the second object;

(e) when the expected time estimate for receiving the second portion of additional autonomy data satisfies a first threshold, determining a predicted updated state of the second object, a predicted track of the first object based on the updated state of the first object, and a predicted track of the second object based on the predicted updated state of the second object; and

(f) controlling navigation of the autonomous vehicle based on the predicted track of the first object and the predicted track of the second object.

2. The autonomy computing system of claim 1 , wherein the interaction classification is indicative of the first object and the second object being within a predetermined distance of one another at one or more of a current time or a future time.

3. The autonomy computing system of claim 1 , wherein (b) comprises accessing an object dependence graph that identifies interacting objects.

4. The autonomy computing system of claim 1 , wherein (b) comprises using a machine-learned model that has been trained based on training data that comprises annotated vehicle data logs that were previously collected during autonomous vehicle driving sessions.

5. The autonomy computing system of claim 1 , the operations comprising:

(g) when the expected time estimate for receiving the second portion of additional autonomy data satisfies a second threshold, obtaining the second portion of additional autonomy data descriptive of the updated state of the second object and determining the predicted track of the second object based on the updated state of the second object.

6. The autonomy computing system of claim 1 , wherein:

one or more of the first portion of additional autonomy data or the second portion of additional autonomy data is obtained by a first system of the autonomy computing system; and

wherein the operations comprise (g) transmitting the predicted track of the first object and the predicted track of the second object to a second system of the autonomy computing system.

7. The autonomy computing system of claim 6 , wherein:

the first system comprises a prediction system configured to determine a predicted track for the first object and the second object; and

the second system comprises a motion planning system configured to determine, for a current time frame, a motion plan for the autonomous vehicle based on track data for the first object and the second object.

8. The autonomy computing system of claim 1 , the operations comprising:

(g) obtaining a third portion of additional autonomy data descriptive of sensor data associated with the first object;

(h) determining an expected time estimate for receiving a fourth portion of additional autonomy data descriptive of sensor data associated with the second object; and

(i) when the expected time estimate for receiving the fourth portion of additional autonomy data satisfies a second threshold, determining a predicted portion of sensor data associated with the second object, an updated state of the first object based on the sensor data associated with the first object, and an updated state of the second object based on the predicted portion of sensor data associated with the second object.

9. The autonomy computing system of claim 8 , wherein:

one or more of the third portion of additional autonomy data or the fourth portion of additional autonomy data is obtained by a first system of the autonomy computing system; and

wherein the operations comprise (j) transmitting the updated state of the first object and the updated state of the second object to a second system of the autonomy computing system.

10. The autonomy computing system of claim 9 , wherein:

the first system comprises a perception system configured to generate state data descriptive of at least a current state of the first object and the second object; and

the second system comprises a prediction system configured to determine, for a current time frame, a predicted track for the first object and the second object.

11. The autonomy computing system of claim 1 , the operations comprising (g) determining a priority classification associated with the first object, the priority classification classifying the first object as one of high-priority or low-priority.

12. The autonomy computing system of claim 11 , wherein determining the predicted updated state of the second object, the predicted track of the first object based on the updated state of the first object and the predicted track of the second object based on the predicted updated state of the second object in (e) is implemented when the priority classification associated with the first object is determined in (g) to be high-priority.

13. An autonomous vehicle (AV) control system, comprising:

one or more processors; and

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

(a) accessing autonomy data descriptive of a state of a first object and a state of a second object in a surrounding environment of an autonomous vehicle;

(b) determining an interaction classification of the first object relative to the second object, the interaction classification indicative of a likelihood that the first object and the second object will interact;

(c) obtaining, by a first system of the AV control system, a first portion of additional autonomy data descriptive of an updated state of the first object;

(d) determining an expected time estimate for receiving a second portion of additional autonomy data descriptive of an updated state of the second object;

(e) when the expected time estimate for receiving the second portion of additional autonomy data satisfies a first threshold, determining a predicted updated state of the second object, a predicted track of the first object based on the updated state of the first object, and a predicted track of the second object based on the predicted updated state of the second object, and transmitting the predicted track of the first object and the predicted track of the second object from the first system to a second system of the AV control system; and

(f) controlling navigation of the autonomous vehicle based on the predicted track of the first object and the predicted track of the second object.

14. The AV control system of claim 13 , the operations comprising (g) determining a priority classification associated with the first object, the priority classification classifying the first object as one of high-priority or low-priority.

15. The AV control system of claim 14 , wherein determining the predicted updated state of the second object, the predicted track of the first object, and the predicted track of the second object in (e) is implemented when the priority classification associated with the first object is determined in (g) to be high-priority.

16. The AV control system of claim 13 , wherein the interaction classification is indicative of the first object and the second object being within a predetermined distance of one another at one or more of a current time or a future time.

17. The AV control system of claim 13 , the operations comprising:

(g) when the expected time estimate for receiving the second portion of additional autonomy data satisfies a second threshold, obtaining the second portion of additional autonomy data descriptive of the updated state of the second object and determining the predicted track of the second object based on the updated state of the second object.

18. The AV control system of claim 13 , wherein:

the first system comprises a prediction system configured to determine a predicted track for the first object and the second object; and

the second system comprises a motion planning system configured to determine, for a current time frame, a motion plan for the autonomous vehicle based on track data for the first object and the second object.

19. An autonomous vehicle, comprising:

one or more processors; and

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

(a) accessing autonomy data descriptive of a state of a first object and a state of a second object in a surrounding environment of an autonomous vehicle;

(b) determining a priority classification associated with the first object, the priority classification classifying the first object as one of high-priority or low-priority;

(c) obtaining a first portion of additional autonomy data descriptive of an updated state of the first object;

(d) when the priority classification associated with the first object is high-priority, determining a predicted updated state for the second object, a predicted track of the first object based on the updated state of the first object, and a predicted track of the second object based on the predicted updated state of the second object; and

(e) controlling navigation of the autonomous vehicle based on the predicted track of the first object and the predicted track of the second object.

20. The autonomous vehicle of claim 19 , wherein:

the first portion of additional autonomy data is obtained by a first system of the autonomous vehicle; and

wherein the operations comprise (f) transmitting the predicted track of the first object and the predicted track of the second object to a second system of the autonomous vehicle.