Systems and methods to control autonomous vehicle motion
The present disclosure provides systems and methods that control the motion of an autonomous vehicle by rewarding or otherwise encouraging progress toward a goal, rather than simply rewarding distance travelled. In particular, the systems and methods of the present disclosure can project a candidate motion plan that describes a proposed motion path for the autonomous vehicle onto a nominal pathway to determine a projected distance associated with the candidate motion plan. The systems and methods of the present disclosure can use the projected distance to evaluate a reward function that provides a reward that is positively correlated to the magnitude of the projected distance. The motion of the vehicle can be controlled based on the reward value provided by the reward function. For example, the candidate motion plan can be selected for implementation or revised based at least in part on the determined reward value.
1. A computer-implemented method to control motion of an autonomous vehicle, the method comprising:
obtaining a candidate motion plan that describes a proposed motion path for the autonomous vehicle;
identifying a nominal pathway associated with the autonomous vehicle, wherein the candidate motion plan deviates from the nominal pathway and wherein the nominal pathway comprises a plurality of nominal pathway locations;
projecting the candidate motion plan onto the nominal pathway to determine a projected distance along the nominal pathway for the candidate motion plan;
selecting the candidate motion plan for implementation based at least in part on the projected distance; and
executing one or more control signals to control the motion of the autonomous vehicle based on the candidate motion plan.
2. The computer-implemented method of claim 1 , wherein the nominal pathway corresponds to a center line of one or more lanes in which the autonomous vehicle intends to travel.
3. The computer-implemented method of claim 1 , wherein projecting the candidate motion plan onto the nominal pathway to determine the projected distance along the nominal pathway for the candidate motion plan, comprises:
determining a plurality of candidate motion plan locations for the candidate motion plan; and
determining the projected distance along the nominal pathway for the candidate motion plan based on a comparison of the plurality of nominal pathway locations to the plurality of candidate motion plan locations.
4. The computer-implemented method of claim 3 , wherein the projected distance along the nominal pathway is based on an L2 distance from at least one of the plurality of candidate motion plan locations to at least one of the nominal pathway locations.
5. The computer-implemented method of claim 4 , wherein determining the projected distance along the nominal pathway for the candidate motion plan comprises:
evaluating a reward function based on the L2 distance from the at least one of the plurality of candidate motion plan locations to the at least one of the nominal pathway locations.
6. The computer-implemented method of claim 1 , further comprising
assigning one or more weights to the candidate motion plan based on the projected distance for the candidate motion plan; and
selecting the candidate motion plan for implementation based at least in part on the one or more weights.
7. The computer-implemented method of claim 6 , wherein the one or more weights are determined based on a scalar parameter configured to control a smoothness of a projection function for projecting the candidate motion plan onto the nominal pathway.
8. The computer-implemented method of claim 1 , wherein the projected distance is determined based on a sum of a plurality of sub-distances.
9. The computer-implemented method of claim 8 , wherein the sub-distances are weighted.
10. The computer-implemented method of claim 8 , wherein the sub-distances are unweighted.
11. A vehicle 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 vehicle control system to perform operations, the operations comprising:
obtaining a candidate motion plan that describes a proposed motion path for an autonomous vehicle;
identifying a nominal pathway associated with the autonomous vehicle, wherein the candidate motion plan deviates from the nominal pathway and wherein the nominal pathway comprises a plurality of nominal pathway locations;
projecting the candidate motion plan onto the nominal pathway to determine a projected distance along the nominal pathway for the candidate motion plan;
selecting the candidate motion plan for implementation based at least in part on the projected distance; and
executing one or more control signals to control the motion of the autonomous vehicle: based on the candidate motion plan.
12. The vehicle control system of claim 11 , wherein the nominal pathway corresponds to a center line of one or more lanes in which the autonomous vehicle intends to travel.
13. The vehicle control system of claim 11 , wherein the projecting the candidate motion plan onto the nominal pathway to determine the projected distance along the nominal pathway for the candidate motion plan, comprises:
determining a plurality of candidate motion plan locations for the candidate motion plan; and
determining the projected distance along the nominal pathway for the candidate motion plan based on a comparison of the plurality of nominal pathway location to the plurality of candidate motion plan locations.
14. The vehicle control system of claim 13 , wherein the projected distance along the nominal pathway is based on an L2 distance from at least one of the plurality of candidate motion plan locations to at least one of the nominal pathway locations.
15. The vehicle control system of claim 14 , wherein determining the projected distance along the nominal pathway for the candidate motion plan comprises:
evaluating a reward function based on the L2 distance from the at least one of the plurality of candidate motion plan locations to the at least one of the nominal pathway locations.
16. An autonomous vehicle, comprising:
one or more processors; and
one or more tangible, non-transitory, computer-readable media that store instructions that, when executed by the one or more processors, cause the one or more processors perform operations, the operations comprising:
obtaining a candidate motion plan that describes a proposed motion path for the autonomous vehicle;
identifying a nominal pathway associated with the autonomous vehicle, wherein the candidate motion plan deviates from the nominal pathway and wherein the nominal pathway comprises a plurality of nominal pathway locations;
projecting the candidate motion plan onto the nominal pathway to determine a projected distance along the nominal pathway for the candidate motion plan;
selecting the candidate motion plan for implementation based at least in part on the projected distance; and
executing one or more control signals to control the motion of the autonomous vehicle based on the candidate motion plan.
17. The autonomous vehicle of claim 16 , further comprising:
assigning one or more weights to the candidate motion plan based on the projected distance for the candidate motion plan; and
selecting the candidate motion plan for implementation based at least in part on the one or more weights.
18. The autonomous vehicle of claim 17 , wherein the one or more weights are determined based on a scalar parameter configured to control a smoothness of a projection function for projecting the candidate motion plan onto the nominal pathway.