US 12,103,560 B2
Methods and systems for predicting actions of an object by an autonomous vehicle to determine feasible paths through a conflicted area
Andrew T. Hartnett, West Hartford, CT (US); Constantin Savtchenko, Sewickley, PA (US); and G. Peter K. Carr, Allison Park, PA (US)
Assigned to Argo AI, LLC, Pittsburgh, PA (US)
Filed by Argo AI, LLC, Pittsburgh, PA (US)
Filed on Oct. 1, 2020, as Appl. No. 17/060,837.
Prior Publication US 2022/0105959 A1, Apr. 7, 2022
Int. Cl. B60W 60/00 (2020.01); B60W 30/18 (2012.01)
CPC B60W 60/0017 (2020.02) [B60W 30/18159 (2020.02); B60W 60/0027 (2020.02); B60W 2554/402 (2020.02); B60W 2554/404 (2020.02)] 28 Claims
OG exemplary drawing
 
1. A method of predicting actions of an actor before entering a conflicted area, the method comprising:
detecting, using sensor data received from one or more sensors of an autonomous vehicle, presence of an actor in an environment of the autonomous vehicle while the autonomous vehicle and the actor are approaching the conflicted area;
determining whether the autonomous vehicle has precedence over the actor for traversing the conflicted area;
in response to determining that the autonomous vehicle has precedence over the actor for traversing the conflicted area, assigning a kinematic target to the actor that requires the actor to come to a stop at a yield point before entering the conflicted area;
at one of a plurality of time steps, in response to determining that the actor cannot achieve the kinematic target, generating, using a motion model, a plurality of forecasted trajectories for the actor; and
controlling, using the plurality of forecasted trajectories, movement of the autonomous vehicle to traverse the conflicted area.