US 12,168,461 B2
Systems and methods for predicting the trajectory of a moving object
Xin Huang, Cambridge, MA (US); Igor Gilitschenski, Newton, MA (US); Guy Rosman, Newton, MA (US); Stephen G. McGill, Jr., Cambridge, MA (US); John Joseph Leonard, Newton, MA (US); Ashkan Mohammadzadeh Jasour, Cambridge, MA (US); and Brian C. Williams, Cambridge, MA (US)
Assigned to Toyota Research Institute, Inc., Los Altos, CA (US); and Massachusetts Institute of Technology, Cambridge, MA (US)
Filed by Toyota Research Institute, Inc., Los Altos, CA (US)
Filed on Dec. 1, 2021, as Appl. No. 17/539,668.
Claims priority of provisional application 63/216,017, filed on Jun. 29, 2021.
Prior Publication US 2022/0410938 A1, Dec. 29, 2022
Int. Cl. B60W 60/00 (2020.01); G05D 1/00 (2024.01); G06N 3/08 (2023.01)
CPC B60W 60/0027 (2020.02) [B60W 60/00256 (2020.02); G05D 1/0212 (2013.01); G06N 3/08 (2013.01); B60W 2555/60 (2020.02); B60W 2556/65 (2020.02)] 20 Claims
OG exemplary drawing
 
1. A system for predicting a trajectory of a moving object, the system comprising:
a processor; and
a memory storing instructions that, when executed by the processor, cause the processor to:
download, to a robot, a probabilistic hybrid discrete-continuous automaton (PHA) model learned as a deep neural network, wherein the learned PHA model models a moving object in an environment of the robot;
use the deep neural network to infer a sequence of high-level discrete modes and a set of associated low-level samples, wherein the high-level discrete modes correspond to candidate maneuvers for the moving object and the low-level samples are candidate trajectories for the moving object;
use the sequence of high-level discrete modes and the set of associated low-level samples, via a learned proposal distribution in the deep neural network, to adaptively sample the sequence of high-level discrete modes to produce a reduced set of low-level samples; and
apply a sample selection technique to the reduced set of low-level samples to select a predicted trajectory for the moving object; and
control operation of the robot based, at least in part, on the predicted trajectory.