Systems and methods for costing autonomous vehicle maneuvers
Systems and methods for autonomous vehicle motion planning are provided. Sensor data describing an environment of an autonomous vehicle and an initial travel path for the autonomous vehicle through the environment can be obtained. A number of trajectories for the autonomous vehicle are generated based on the sensor data and the initial travel path. The trajectories can be evaluated by generating a number of costs for each trajectory. The costs can include a safety cost and a total cost. Each cost is generated by a cost function created in accordance with a number of relational propositions defining desired relationships between the number of costs. A subset of trajectories can be determined from the trajectories based on the safety cost and an optimal trajectory can be determined from the subset of trajectories based on the total cost. The autonomous vehicle can control its motion in accordance with the optimal trajectory.
1. A computing system, the computing system comprising:
one or more processors; and
one or more tangible, non-transitory, computer readable media that collectively store instructions that when executed by the one or more processors cause the computing system to perform operations, the operations comprising:
obtaining a plurality of sub-cost functions configured to evaluate a trajectory for controlling a motion of an autonomous vehicle, wherein each sub-cost function is configured to generate a respective sub-cost value indicative of the trajectory's impact on one or more motion planning criteria, wherein each sub-cost function is configured to generate the respective sub-cost value based, at least in part, on sensor data indicative of a surrounding environment of the autonomous vehicle, a period of time, and the trajectory;
obtaining one or more relational propositions associated with the plurality of sub-cost functions, wherein each relational proposition is indicative of a desired relationship between at least two of the plurality of sub-cost functions;
generating verification data indicative of whether the plurality of sub-cost functions achieve the one or more relational propositions based, at least in part, on the plurality of sub-cost functions and the one or more relational propositions; and
providing for display, via one or more display devices, the verification data.
2. The computing system of claim 1 , wherein one or more motion planning criteria comprise one or more safety criteria and one or more secondary criteria, wherein the one or more safety criteria are indicative of interference avoidance, road departure avoidance, or lane departure avoidance.
3. The computing system of claim 2 , wherein the plurality of sub-cost functions comprise one or more safety sub-cost functions and one or more secondary sub-cost functions, wherein the one or more safety sub-cost functions are configured to generate one or more safety sub-cost values indicative of the trajectory's impact on at least one of the one or more safety criteria, and wherein the one or more secondary sub-cost functions are configured to generate one or more secondary sub-cost values indicative of the trajectory's impact on at least one of the one or more secondary criteria.
4. The computing system of claim 3 , wherein at least one of the one or more relational propositions identifies a safety relationship between each of the one or more safety sub-cost functions and the one or more secondary sub-cost functions, wherein the safety relationship defines a requirement that each of the one or more safety sub-cost values generated by the one or more safety sub-cost functions outweigh any combination of the one or more secondary sub-cost values generated by the one or more secondary sub-cost functions.
5. The computing system of claim 1 , wherein the operations further comprise: generating one or more cost modifications for the plurality of sub-cost functions based, at least in part, on the one or more relational propositions.
6. The computing system of claim 5 , wherein the verification data indicates that the plurality of sub-cost functions do not achieve the one or more relational propositions, and wherein the operations further comprise:
determining at least one deviating relational proposition, the at least one deviating relational proposition indicative of at least one of the one or more relational propositions that is not achieved by the plurality of sub-cost functions; and
generating the one or more cost modifications for the plurality of sub-cost functions based, at least in part, on the at least one deviating relational proposition.
7. The computing system of claim 5 , wherein the operations further comprise:
obtaining user input indicative of a costing change to at least one of the plurality of sub-cost functions or the one or more relational propositions; and
generating the verification data based, at least in part, on the costing change.
8. The computing system of claim 7 , wherein the costing change comprises an additional sub-cost function in addition to the plurality of sub-cost functions, and wherein the operations further comprise:
generating the one or more cost modifications for the plurality of sub-cost functions based, at least in part, on the additional sub-cost function.
9. The computing system of claim 7 , wherein the costing change comprises an additional relational proposition in addition to the one or more relational propositions, and wherein the operations further comprise:
generating the one or more cost modifications for the plurality of sub-cost functions based, at least in part, on the additional relational proposition.
10. A computing system, the computing system comprising:
one or more processors; and
one or more tangible, non-transitory, computer readable media that collectively store instructions that when executed by the one or more processors cause the computing system to perform operations, the operations comprising:
obtaining sensor data descriptive of an environment of an autonomous vehicle;
evaluating a plurality of sub-cost functions for a trajectory of the autonomous vehicle, wherein each sub-cost function is configured to generate a respective sub-cost value indicative of the trajectory's impact on one or more motion planning criteria, wherein each sub-cost function is configured to generate the respective sub-cost value based, at least in part, on the sensor data descriptive of the environment of the autonomous vehicle, a period of time, and the trajectory;
determining that the respective sub-cost values for the trajectory satisfy sub-cost criteria;
determining that one or more relational propositions associated with the plurality of sub-cost functions for the trajectory are achieved, wherein each relational proposition is indicative of a desired relationship between at least two of the plurality of sub-cost functions; and
controlling the autonomous vehicle in accordance with the trajectory.
11. The computing system of claim 10 , wherein one or more motion planning criteria comprise one or more safety criteria and one or more secondary criteria, wherein the one or more safety criteria are indicative of interference avoidance, road departure avoidance, or lane departure avoidance.
12. The computing system of claim 11 , wherein the plurality of sub-cost functions comprise one or more safety sub-cost functions and one or more secondary sub-cost functions, wherein the one or more safety sub-cost functions are configured to generate one or more safety sub-cost values indicative of the trajectory's impact on at least one of the one or more safety criteria, and wherein the one or more secondary sub-cost functions are configured to generate one or more secondary sub-cost values indicative of the trajectory's impact on at least one of the one or more secondary criteria.
13. The computing system of claim 12 , wherein at least one of the one or more relational propositions identifies a safety relationship between each of the one or more safety sub-cost functions and the one or more secondary sub-cost functions, wherein the safety relationship defines a requirement that each of the one or more safety sub-cost values generated by the one or more safety sub-cost functions outweigh any combination of the one or more secondary sub-cost values generated by the one or more secondary sub-cost functions.