Motion planning system of an autonomous vehicle
Generally, the present disclosure is directed to systems and methods that include or otherwise leverage a decision point engine as part of determining a motion plan for an autonomous vehicle. In particular, a motion planning system that includes a decision point engine can be configured to obtain object data associated with one or more objects identified near one or more travel paths of an autonomous vehicle. The system can determine a stopping profile based at least in part on the object data, wherein the stopping profile identifies a set of candidate states for the autonomous vehicle that preserves an ability of the autonomous vehicle to stop before the interaction point. The system can determine a decision point corresponding to a selected state from the set of candidate states identified by the stopping profile, and determine a trajectory for the autonomous vehicle based at least in part on the decision point.
1. A computing system for autonomous vehicle control, the 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 computing system to perform operations, the operations comprising:
obtaining object data associated with an object identified near one or more travel paths of an autonomous vehicle, wherein the object data comprises at least a probability of interaction between the autonomous vehicle and the object at an interaction point;
determining a stopping profile based at least in part on the object data, wherein the stopping profile identifies a set of candidate states for the autonomous vehicle that preserves an ability of the autonomous vehicle to stop before the interaction point;
determining a decision point corresponding to a selected state from the set of candidate states identified by the stopping profile;
generating a speed zone constraint based at least in part on the stopping profile and the decision point that defines an area within a multi-dimensional space that limits a travel speed of the autonomous vehicle;
determining an optimum speed profile from a plurality of speed profiles for implementation by the autonomous vehicle, wherein each of the plurality of speed profiles is determined to satisfy the speed zone constraint and a plurality of other constraints associated with travel by the autonomous vehicle;
determining a trajectory for the autonomous vehicle based at least in part on the optimum speed profile for implementation by the autonomous vehicle; and
controlling motion of the autonomous vehicle based at least in part on the trajectory.
2. The computing system of claim 1 , wherein determining the decision point comprises determining the selected state from the set of candidate states based at least in part on where the stopping profile intersects with a linear representation of an estimated time of arrival of the autonomous vehicle at the interaction point.
3. The computing system of claim 1 , wherein the speed zone constraint and the plurality of other constraints are respectively defined as a constraint area in the multi-dimensional space that identifies regions along the one or more travel paths that the autonomous vehicle cannot occupy at certain times, wherein the multi-dimensional space comprises at least a time dimension and a distance dimension relative to travel along the one or more travel paths.
4. The computing system of claim 1 , wherein the stopping profile is determined based at least in part on the probability of interaction between the autonomous vehicle and the object at the interaction point.
5. The computing system of claim 1 , wherein determining the stopping profile based at least in part on the object data is implemented when the probability of interaction between the autonomous vehicle and the object is greater than a predetermined threshold probability value.
6. The computing system of claim 1 , wherein the stopping profile is determined based at least in part on a maximum deceleration value and a maximum jerk value, wherein the maximum deceleration value is indicative of a maximum level of force applied to a braking actuator associated with the autonomous vehicle, and wherein the maximum jerk value is indicative of a maximum rate of change for achieving the maximum deceleration value.
7. The computing system of claim 1 , wherein the object data including the probability of interaction between the autonomous vehicle and the object at the interaction point is determined from travel path data descriptive of the one or more travel paths and prediction data descriptive of one or more predicted future locations for identified objects proximate to the autonomous vehicle.
8. The computing system of claim 7 , wherein the travel path data is descriptive of one or more predetermined potential travel paths of the autonomous vehicle based at least in part on a common pattern of vehicle travel along one or more lanes of a roadway.
9. A computer-implemented method, comprising:
obtaining, object data associated with one or more objects identified near one or more travel paths of an autonomous vehicle, wherein the object data comprises at least a probability of interaction between the autonomous vehicle and at least one of the objects at an interaction point;
determining, a stopping profile based at least in part on the object data, wherein the stopping profile identifies a set of candidate states for the autonomous vehicle that preserves an ability of the autonomous vehicle to stop before the interaction point;
determining, a decision point corresponding to a selected state from the set of candidate states identified by the stopping profile, wherein the decision point is determined based at least in part on where the stopping profile intersects with a linear representation of an estimated arrival time of the autonomous vehicle at the interaction point;
generating a speed zone constraint based at least in part on the stopping profile and the decision point that defines an area within a multi-dimensional space that limits a travel speed of the autonomous vehicle;
determining an optimum speed profile from a plurality of speed profiles for implementation by the autonomous vehicle, wherein each of the plurality of speed profiles is determined to satisfy the speed zone constraint and a plurality of other constraints associated with travel by the autonomous vehicle;
determining, by the computing system, a trajectory for the autonomous vehicle based at least in part on the optimum speed profile for implementation by the autonomous vehicle; and
controlling, motion of the autonomous vehicle based at least in part on the trajectory.
10. The computer-implemented method of claim 9 , wherein the speed zone constraint and the plurality of other constraints are respectively defined as a constraint area in the multi-dimensional space that identifies regions along the one or more travel paths that the autonomous vehicle cannot occupy at certain times, wherein the multi-dimensional space comprises at least a time dimension and a distance dimension relative to travel along the one or more travel paths.
11. The computer-implemented method of claim 9 , wherein the stopping profile is determined based at least in part on the probability of interaction between the autonomous vehicle and the at least one of the objects at the interaction point.
12. The computer-implemented method of claim 9 , wherein determining the stopping profile based at least in part on the object data is implemented when the probability of interaction between the autonomous vehicle and the at least one of the objects is greater than a predetermined threshold probability value.
13. The computer-implemented method of claim 9 , wherein the stopping profile is determined based at least in part on a maximum deceleration value and a maximum jerk value, wherein the maximum deceleration value is indicative of a maximum level of force applied to a braking actuator associated with the autonomous vehicle, and wherein the maximum jerk value is indicative of a maximum rate of change for achieving the maximum deceleration value.
14. The computer-implemented method of claim 9 , wherein the object data including the probability of interaction between the autonomous vehicle and the at least one of the objects at the interaction point is determined from travel path data and prediction data, wherein the travel path data is descriptive of one or more predetermined potential paths for travel of the autonomous vehicle based at least in part on a common pattern of vehicle travel along one or more lanes of a roadway, and wherein the prediction data is descriptive of one or more predicted future locations for identified objects proximate to the autonomous vehicle.
15. 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 object data associated with one or more objects identified near one or more travel paths of the autonomous vehicle, wherein the object data comprises at least a probability of interaction between the autonomous vehicle and at least one of the one or more objects at an interaction point;
determining a stopping profile based at least in part on the object data, wherein the stopping profile identifies a set of candidate states for the autonomous vehicle that preserves an ability of the autonomous vehicle to stop before the interaction point;
determining a decision point corresponding to a selected state from the set of candidate states identified by the stopping profile;
generating a speed zone constraint based at least in part on the stopping profile and the decision point that defines an area within a multi-dimensional space that limits travel speed of the autonomous vehicle; and
determining an optimum speed profile from a plurality of speed profiles for implementation by the autonomous vehicle, wherein each of the plurality of speed profiles is determined to satisfy the speed zone constraint and a plurality of other constraints associated with travel by the autonomous vehicle;
determining a trajectory based at least in part on the optimum speed profile for implementation by the autonomous vehicle; and
controlling motion of the autonomous vehicle based at least in part on the trajectory.
16. The autonomous vehicle of claim 15 , wherein determining the decision point comprises determining the selected state from the set of candidate states based at least in part on where the stopping profile intersects with a linear representation of an estimated time of arrival of the autonomous vehicle at the interaction point.
17. The autonomous vehicle of claim 15 , wherein the stopping profile is determined based at least in part on a maximum deceleration value and a maximum jerk value, wherein the maximum deceleration value is indicative of a maximum level of force applied to a braking actuator associated with the autonomous vehicle, and wherein the maximum jerk value is indicative of a maximum rate of change for achieving the maximum deceleration value.
18. The autonomous vehicle of claim 15 , wherein the object data including the probability of interaction between the autonomous vehicle and the at least one object at the interaction point is determined from travel path data and prediction data, wherein the travel path data is descriptive of one or more predetermined potential paths for travel of the autonomous vehicle based at least in part on a common pattern of vehicle travel along one or more lanes of a roadway, and wherein the prediction data is descriptive of one or more predicted future locations for identified objects proximate to the autonomous vehicle.