US 12,420,791 B2
Autonomous vehicle prediction layer training
Thanard Kurutach, Carrboro, NC (US); and Ariel Arturo Perez Chavez, Mountain View, CA (US)
Assigned to GM Cruise Holdings LLC, San Francisco, CA (US)
Filed by GM Cruise Holdings LLC, San Francisco, CA (US)
Filed on Jan. 23, 2023, as Appl. No. 18/158,424.
Prior Publication US 2024/0246537 A1, Jul. 25, 2024
Int. Cl. B60W 30/16 (2020.01); B60W 50/00 (2006.01); B60W 60/00 (2020.01); G06N 3/08 (2023.01)
CPC B60W 30/16 (2013.01) [B60W 50/0097 (2013.01); B60W 60/0011 (2020.02); B60W 60/0015 (2020.02); B60W 60/00274 (2020.02); G06N 3/08 (2013.01); B60W 2420/408 (2024.01); B60W 2420/54 (2013.01); B60W 2554/4045 (2020.02); B60W 2554/80 (2020.02); B60W 2556/40 (2020.02); B60W 2556/50 (2020.02)] 17 Claims
OG exemplary drawing
 
1. A system comprising:
at least one memory; and
at least one processor coupled to the at least one memory, the at least one processor configured to:
receive road data, wherein the road data represents a real-world environment encountered by an autonomous vehicle (AV);
generate, using a prediction layer of the AV, a predicted trajectory of a target vehicle, wherein the predicted trajectory comprises one or more waypoints and wherein the predicted trajectory is based on the road data;
calculate a distance metric for the predicted trajectory, wherein the distance metric is based on a distance between the one or more waypoints and one or more corresponding drivable areas, wherein the one or more corresponding drivable areas are nearest drivable areas to the one or more waypoints;
update the prediction layer of the AV based on the distance metric; and
control navigation of the autonomous vehicle using subsequent trajectory predictions generated by the updated prediction layer to avoid navigating into non-drivable areas.