Systems and methods for detecting surprise movements of an actor with respect to an autonomous vehicle
Systems and methods for detecting a surprise or unexpected movement of an actor with respect to an autonomous vehicle are provided. An example computer-implemented method can include, for a first compute cycle, obtaining motion forecast data based on first sensor data collected with respect to an actor relative to an autonomous vehicle; and determining, based on the motion forecast data, failsafe region data representing an unexpected path or area where a likelihood of the actor following the unexpected path or entering the unexpected area is below a threshold. For a second compute cycle after the first compute cycle, the method can include obtaining second sensor data; determining, based on the second sensor data and the failsafe region data, that the actor has followed the unexpected path or entered the unexpected area; and in response to such determination, determining a deviation for controlling a movement of the autonomous vehicle.
1. An autonomous vehicle control system for controlling an autonomous vehicle, the autonomous vehicle control system comprising:
one or more processors; and
one or more non-transitory computer-readable media storing instructions that are executable to cause the one or more processors to perform operations, the operations comprising:
obtaining sensor data descriptive of an actor in an environment of an autonomous vehicle;
determining, based on the sensor data, a plurality of foreseeable paths for the actor;
determining, in a first compute cycle, and based on the plurality of foreseeable paths, a preliminary action for controlling a movement of the autonomous vehicle, the preliminary action configured for execution as a precaution to improve a reaction of the autonomous vehicle to the actor traveling on one of the plurality of foreseeable paths; and
initiating control of the movement of the autonomous vehicle according to the preliminary action by executing the preliminary action at least until a second compute cycle.
2. The autonomous vehicle control system of claim 1 , wherein the operations comprise:
stopping execution of the preliminary action based on a determination that the actor did not travel on the one of the plurality of foreseeable paths.
3. The autonomous vehicle control system of claim 1 , wherein the preliminary action diverges from a motion plan generated based on a determination that the actor will not travel on the one of the plurality of foreseeable paths.
4. The autonomous vehicle control system of claim 1 , wherein the preliminary action comprises a braking action.
5. The autonomous vehicle control system of claim 1 , wherein initiating control of the movement of the autonomous vehicle comprises:
controlling one or more actuators of the autonomous vehicle based on the preliminary action.
6. The autonomous vehicle control system of claim 1 , wherein the first compute cycle corresponds to processing a first LIDAR sweep and the second compute cycle corresponds to processing a subsequent LIDAR sweep.
7. The autonomous vehicle control system of claim 1 , wherein the operations comprise:
determining a first motion plan and a second motion plan, wherein the second motion plan corresponds to the reaction; and
executing the first motion plan after execution of the preliminary action.
8. The autonomous vehicle control system of claim 1 , wherein the operations comprise:
determining a first motion plan and a second motion plan, wherein the second motion plan corresponds to the reaction; and
executing the second motion plan after execution of the preliminary action and responsive to the actor traveling on one of the plurality of foreseeable paths.
9. One or more non-transitory computer-readable media storing instructions that are executable to cause one or more processors to perform operations for controlling an autonomous vehicle, the operations comprising:
obtaining sensor data descriptive of an actor in an environment of an autonomous vehicle;
determining, based on the sensor data, a plurality of foreseeable paths for the actor;
determining, in first compute cycle, and based on the plurality of foreseeable paths, a preliminary action for controlling a movement of the autonomous vehicle, the preliminary action configured for execution as a precaution to improve a reaction of the autonomous vehicle to the actor traveling on one of the plurality of foreseeable paths; and
initiating control of the movement of the autonomous vehicle according to the preliminary action by executing the preliminary action at least until a second compute cycle.
10. The one or more non-transitory computer-readable media of claim 9 , wherein the operations comprise:
stopping execution of the preliminary action based on a determination that the actor did not travel on the one of the plurality of foreseeable paths.
11. The one or more non-transitory computer-readable media of claim 9 , wherein the preliminary action diverges from a motion plan determined under an assumption that the actor will not travel on one of the plurality of foreseeable paths.
12. The one or more non-transitory computer-readable media of claim 9 , wherein the preliminary action comprises a braking action.
13. The one or more non-transitory computer-readable media of claim 9 , wherein initiating control of the movement of the autonomous vehicle comprises:
controlling one or more actuators of the autonomous vehicle based on the preliminary action.
14. The one or more non-transitory computer-readable media of claim 9 , wherein the first compute cycle corresponds to processing a first LIDAR sweep and the second compute cycle corresponds to processing a subsequent LIDAR sweep.
15. The one or more non-transitory computer-readable media of claim 9 , wherein the operations comprise:
determining a first motion plan and a second motion plan, wherein the second motion plan corresponds to the reaction; and
executing the first motion plan after execution of the preliminary action.
16. The one or more non-transitory computer-readable media of claim 9 , wherein the operations comprise:
determining a first motion plan and a second motion plan, wherein the second motion plan corresponds to the reaction; and
executing the second motion plan responsive to the actor traveling on one of the plurality of foreseeable paths and after execution of the preliminary action.
17. A method for controlling an autonomous vehicle, the method comprising:
obtaining sensor data descriptive of an actor in an environment of an autonomous vehicle;
determining, based on the sensor data, a plurality of foreseeable paths for the actor;
determining, in first compute cycle, and based on the plurality of foreseeable paths, a preliminary action for controlling a movement of the autonomous vehicle, the preliminary action configured for execution as a precaution to improve a reaction of the autonomous vehicle to the actor traveling on one of the plurality of foreseeable paths; and
initiating control of the movement of the autonomous vehicle according to the preliminary action by executing the preliminary action at least until a second compute cycle.
18. The method of claim 17 , comprising:
stopping execution of the preliminary action based on a determination that the actor did not travel on the one of the plurality of foreseeable paths.
19. The method of claim 17 , comprising:
determining a first motion plan and a second motion plan, wherein the second motion plan corresponds to the reaction; and
executing the second motion plan responsive to the actor traveling on one of the plurality of foreseeable paths and after execution of the preliminary action.
20. The method of claim 17 , wherein the first compute cycle corresponds to processing a first LIDAR sweep and the second compute cycle corresponds to processing a subsequent LIDAR sweep.