US 12,391,246 B2
Trajectory optimization in multi-agent environments
Marin Kobilarov, Baltimore, MD (US); and Chonhyon Park, San Jose, CA (US)
Assigned to Zoox, Inc., Foster City, CA (US)
Filed by Zoox, Inc., Foster City, CA (US)
Filed on Aug. 31, 2022, as Appl. No. 17/900,258.
Prior Publication US 2024/0092357 A1, Mar. 21, 2024
Int. Cl. B60W 30/095 (2012.01)
CPC B60W 30/0956 (2013.01) [B60W 2555/60 (2020.02)] 20 Claims
OG exemplary drawing
 
1. A vehicle comprising:
one or more processors; and
one or more non-transitory computer-readable media storing computer-executable instructions that, when executed, cause the one or more processors to perform operations comprising:
receiving a vehicle trajectory associated with controlling the vehicle in an environment including the vehicle and a dynamic object different from the vehicle, wherein the vehicle trajectory comprises a plurality of trajectory points;
determining, for a trajectory point of the plurality of trajectory points, vector representative of a vehicle state value associated with the trajectory point;
perturbing the vector into a perturbed vector, the perturbed vector representing a portion of a perturbed vehicle trajectory associated with the vehicle;
determining, based at least in part on the perturbed vehicle trajectory, a kino-dynamically feasible perturbed trajectory for the vehicle;
inputting the kino-dynamically feasible perturbed trajectory and a representation of the environment into a machine-learned model;
receiving, from the machine-learned model, a predicted object trajectory for the dynamic object;
determining, based at least in part on the kino-dynamically feasible perturbed trajectory for the vehicle and the predicted object trajectory, a cost associated with the kino-dynamically feasible perturbed trajectory;
determining the kino-dynamically feasible perturbed trajectory as a control trajectory for the vehicle, based at least in part on the cost; and
controlling the vehicle in the environment, based at least in part on the control trajectory.