US 11,868,137 B2
Systems and methods for path planning with latent state inference and graphical relationships
Jiachen Li, Albany, CA (US); David F. Isele, San Jose, CA (US); Kikuo Fujimura, Palo Alto, CA (US); Xiaobai Ma, Stanford, CA (US); and Mykel J. Kochenderfer, Palo Alto, CA (US)
Assigned to HONDA MOTOR CO., LTD., Tokyo (JP)
Filed by Honda Motor Co., Ltd., Tokyo (JP); and The Board of Trustees of the Leland Stanford Junior University, Stanford, CA (US)
Filed on Feb. 11, 2021, as Appl. No. 17/173,753.
Claims priority of provisional application 63/113,146, filed on Nov. 12, 2020.
Prior Publication US 2022/0147051 A1, May 12, 2022
Int. Cl. G06N 3/08 (2023.01); B60W 40/04 (2006.01); G05D 1/02 (2020.01)
CPC G05D 1/0221 (2013.01) [B60W 40/04 (2013.01); G05D 1/0219 (2013.01); G05D 1/0274 (2013.01); G06N 3/08 (2013.01); B60W 2552/10 (2020.02); B60W 2555/60 (2020.02); G05D 2201/0213 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A system for path planning with latent state inference and graphical relationships, the system comprising:
an inference module configured to:
receive sensor data associated with a plurality of agents, wherein the plurality of agents includes an ego agent; and
map the sensor data to a latent state distribution to identify latent states of the plurality of agents, wherein the latent states identify agents of the plurality of agents as cooperative or aggressive;
a policy module configured to predict future trajectories of the plurality of agents at a given time based on the sensor data and the latent states of the plurality of agents;
a graphical representation module configured generate a graphical representation based on the sensor data and a graphical representation neural network, wherein temporal information is applied to the sensor data so the graphical representation includes spatial-temporal relationships between the ego agent and the agents of the plurality of agents;
a planning module configured to generate a motion plan for the ego agent based on the predicted future trajectories and the graphical representation; and
an execution module configured to cause the ego agent to execute the motion plan.