US 12,448,001 B2
Autonomous vehicle motion planning
J. Andrew Bagnell, Pittsburgh, PA (US); Michael William Bode, Pittsburgh, PA (US); Micol Marchetti-Bowick, Pittsburgh, PA (US); Sanjiban Choudry, Ithaca, NY (US); Pengju Jin, San Francisco, CA (US); Sumit Kumar, Sunnyvale, CA (US); Yuhang Ma, Pittsburgh, PA (US); Venkatraman Narayanan, Mountain View, CA (US); Arun Venkatraman, Mountain View, CA (US); and Carl Wellington, Pittsburgh, PA (US)
Assigned to AURORA OPERATIONS, INC., Pittsburgh, PA (US)
Filed by Aurora Operations, Inc., Pittsburgh, PA (US)
Filed on Oct. 1, 2024, as Appl. No. 18/903,361.
Claims priority of provisional application 63/616,284, filed on Dec. 29, 2023.
Prior Publication US 2025/0214618 A1, Jul. 3, 2025
Int. Cl. B60W 60/00 (2020.01); B60W 50/00 (2006.01); G05B 13/02 (2006.01)
CPC B60W 60/0011 (2020.02) [B60W 50/0097 (2013.01); B60W 60/00272 (2020.02); G05B 13/027 (2013.01); B60W 2554/4045 (2020.02)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method, comprising:
(a) obtaining context data descriptive of an environment surrounding an autonomous vehicle, the context data based on map data and perception data;
(b) generating, by a proposer and based on the context data:
(i) a plurality of candidate trajectories, and
(ii) a plurality of actor forecasts for a plurality of actors in the environment;
(c) generating, by a ranker and based on the context data, the plurality of candidate trajectories, and the plurality of actor forecasts, a ranking of the plurality of candidate trajectories; and
(d) controlling a motion of the autonomous vehicle based on a candidate trajectory selected based on the ranking of the plurality of candidate trajectories,
wherein the proposer uses a first machine-learned model to generate the plurality of actor forecasts, and the ranker uses a second machine-learned model to generate the ranking, and wherein:
the first machine-learned model and the second machine-learned model use a common backbone architecture;
the first machine-learned model comprises a first decoder that processes outputs of the common backbone architecture;
the second machine-learned model comprises a second decoder that processes outputs of the common backbone architecture and the plurality of candidate trajectories;
the first machine-learned model uses a first backbone model characterized by the common backbone architecture;
the second machine-learned model uses a second backbone model characterized by the common backbone architecture; and
the first backbone model and the second backbone model are parameterized by different sets of learned weights.