Autonomous vehicle safe stop

1. A computer-implemented method of autonomous vehicle operation, the computer-implemented method comprising:

receiving, by a computing system comprising one or more computing devices, state data comprising information associated with at least one of: one or more states of an autonomous vehicle or one or more states of an environment external to the autonomous vehicle;

determining, by the computing system, that one or more vehicle stoppage conditions are satisfied based at least in part on the state data, wherein each of the one or more vehicle stoppage conditions is a condition that indicates that the autonomous vehicle is to stop traveling;

selecting, by the computing system based at least in part on the state data and a machine-learned model, a severity level for the one or more satisfied vehicle stoppage conditions from a plurality of available severity levels,

wherein the machine-learned model is trained to select the severity level based at least in part on the one or more satisfied vehicle stoppage conditions, the severity level indicating an immediacy with which the autonomous vehicle is to stop traveling;

generating, by the computing system, a motion plan based at least in part on the severity level; and

controlling, by the computing system, autonomous driving of the autonomous vehicle in accordance with the motion plan.

2. The computer-implemented method of claim 1 , wherein selecting the severity level comprises:

determining, by the computing system, one or more motion characteristics of the autonomous vehicle based at least in part on the state data, the one or more motion characteristics comprising at least one of a velocity, an acceleration, or a trajectory of the autonomous vehicle.

3. The computer-implemented method of claim 1 , wherein the severity level is inversely proportional to at least one of: a distance between a current location of the autonomous vehicle and a destination location, an amount of change in a velocity of the autonomous vehicle, an amount of change in an acceleration of the autonomous vehicle, or an amount of change in a trajectory of the autonomous vehicle.

4. The computer-implemented method of claim 1 , wherein generating the motion plan comprises:

generating the motion plan for the autonomous vehicle to comply with one or more constraints associated with the severity level.

5. The computer-implemented method of claim 4 , wherein generating the motion plan for the autonomous vehicle to comply with the one or more constraints associated with the severity level comprises:

determining, by the computing system, whether the autonomous vehicle is capable of traversing, from a current location within one or more time intervals, at least one of a plurality of paths based at least in part on the state data, the traversal comprising stopping at a terminal location of each of the at least one of the plurality of paths, respectively.

6. The computer-implemented method of claim 5 , wherein generating the motion plan to comply with the one or more constraints associated with the severity level comprises:

determining, by the computing system, a highest ranked path of the plurality of paths based at least in part on one or more path criteria associated with one or more adverse conditions for the autonomous vehicle to avoid.

7. The computer-implemented method of claim 1 , wherein generating the motion plan comprises:

determining, by the computing system, a path from among a plurality of paths with a minimum deviation.

8. The computer-implemented method of claim 7 , wherein the minimum deviation comprises a least amount of change to a velocity, an acceleration, or a trajectory of the autonomous vehicle.

9. The computer-implemented method of claim 1 , wherein generating the motion plan comprises:

in response to at least one of a plurality of paths not being obstructed, determining, by the computing system, a path of the plurality of paths that comprises an area of predetermined size at which the vehicle can stop.

10. The computer-implemented method of claim 1 , wherein generating the motion plan comprises:

constraining, by the computing system, a search space over which the autonomous vehicle optimizes a cost function.

11. The computer-implemented method of claim 1 , wherein generating the motion plan comprises:

modifying, by the computing system, a destination location to comply with one or more constraints associated with the severity level; and

determining, by the computing system, a path of the autonomous vehicle to the destination location.

12. The computer-implemented method of claim 1 , wherein the motion plan comprises stopping the autonomous vehicle without changing a trajectory of the autonomous vehicle within a first threshold time period or a first threshold distance, changing the trajectory of the autonomous vehicle and stopping the autonomous vehicle before a second time period or a second threshold distance, or changing the trajectory of the autonomous vehicle and stopping the autonomous vehicle after the second time period or the second threshold distance.

13. A computing system, comprising: one or more processors; and

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

receiving state data comprising information associated with at least one of: one or more states of an autonomous vehicle or one or more states of an environment external to the autonomous vehicle;

determining that one or more vehicle stoppage conditions are satisfied based at least in part on the state data, wherein each of the one or more vehicle stoppage conditions is a condition that indicates that the autonomous vehicle is to stop traveling;

selecting, based at least in part on the state data and a machine-learned model, a severity level for the one or more satisfied vehicle stoppage conditions from a plurality of available severity levels,

wherein the machine-learned model is trained to select the severity level based at least in part on the one or more satisfied vehicle stoppage conditions, the severity level indicating an immediacy with which the autonomous vehicle is to stop traveling;

generating a motion plan based at least in part on the severity level; and

controlling autonomous driving of the autonomous vehicle in accordance with the motion plan.

14. The computing system of claim 13 , wherein the selected severity level is associated with a surface on which the autonomous vehicle is to stop.

15. The computing system of claim 13 , wherein generating the motion plan comprises:

generating the motion plan for the autonomous vehicle to comply with one or more constraints associated with the severity level.

16. The computing system of claim 15 , wherein generating the motion plan for the autonomous vehicle to comply with one or more constraints associated with the severity level comprises:

modifying a destination location of the autonomous vehicle to comply with the one or more constraints associated with the severity level.

17. The computing system of claim 13 , wherein generating the motion plan comprises:

determining a path, of a plurality of paths, for the autonomous vehicle to follow to stop the autonomous vehicle.

18. The computing system of claim 13 , wherein the one or more states of the environment external to the autonomous vehicle are based at least in part on sensor data received via one or more sensors of the autonomous vehicle.

19. The computing system of claim 13 , wherein the one or more states of the environment external to the autonomous vehicle comprise at least one of a location, a size, or a position of an object within the environment.

20. An autonomous vehicle, comprising:

one or more processors; and

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

receiving state data comprising information associated with at least one of: one or more states of an autonomous vehicle or one or more states of an environment external to the autonomous vehicle;

determining that one or more vehicle stoppage conditions are satisfied based at least in part on the state data, wherein each of the one or more vehicle stoppage conditions is a condition that indicates that the autonomous vehicle is to stop traveling;

selecting, based at least in part on the state data and a machine-learned model, a severity level for the one or more vehicle stoppage conditions from a plurality of available severity levels,

wherein the machine-learned model is trained to select the severity level based at least in part on the one or more satisfied vehicle stoppage conditions, the severity level indicating an immediacy with which the autonomous vehicle is to stop traveling;

generating a motion plan based at least in part on the severity level; and

controlling autonomous driving of the autonomous vehicle in accordance with the motion plan.

21. The autonomous vehicle of claim 20 , wherein generating the motion plan comprises:

modifying a destination location to comply with one or more constraints associated with the severity level; and

determining a path of the autonomous vehicle to the destination location.

22. The autonomous vehicle of claim 21 , further comprising:

implementing the motion plan for the autonomous vehicle to traverse the path.