US 11,987,269 B2
Safe non-conservative planning for autonomous vehicles
Chris L. Baker, Pittsburgh, PA (US); Hung-Jui Huang, Cambridge, MA (US); Yibiao Zhao, Cambridge, MA (US); Michal Cap, Watertown, MA (US); and Kai-Chi Huang, Cambridge, MA (US)
Assigned to isee, Cambridge, MA (US)
Filed by isee, Cambridge, MA (US)
Filed on Oct. 29, 2021, as Appl. No. 17/514,197.
Claims priority of provisional application 63/107,958, filed on Oct. 30, 2020.
Prior Publication US 2022/0135076 A1, May 5, 2022
Int. Cl. B60W 60/00 (2020.01); B60W 50/14 (2020.01); G01C 21/34 (2006.01)
CPC B60W 60/0015 (2020.02) [B60W 50/14 (2013.01); B60W 60/0011 (2020.02); G01C 21/343 (2013.01)] 45 Claims
OG exemplary drawing
 
1. One or more non-transitory computer-readable media storing instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising:
obtaining a first risk budget constraining a first plan for an autonomous vehicle to satisfy a first objective;
based at least on the first risk budget and the first objective, planning a first planned trajectory, over a period of time including a plurality of timesteps, of the autonomous vehicle toward the first objective, at least by:
(a) determining a first risk cost associated with a first initial planned action to be executed over a first timestep of the plurality of timesteps, the first initial planned action corresponding to the first planned trajectory,
(b) based at least on the first risk cost, determining whether the first planned trajectory is feasible or infeasible within the first risk budget, the first planned trajectory being feasible if a predicted amount of risk associated with the first planned trajectory is less than a first remaining risk budget, and the first planned trajectory being infeasible if the predicted amount of risk is greater than the first remaining risk budget, and
(c) responsive to determining that the first planned trajectory is feasible within the first risk budget, executing the first initial planned action over the first timestep of the first planned trajectory, wherein the executing of the first initial planned action comprises applying a control to the autonomous vehicle;
decreasing the first risk budget by the first risk cost associated with the first initial planned action, to obtain the first remaining risk budget;
obtaining first state data corresponding to a state of the autonomous vehicle after executing the first initial planned action of the first planned trajectory; and
based at least on the first state data, the first remaining risk budget, and the first objective, planning a second trajectory of the autonomous vehicle toward the first objective.