US 12,233,887 B2
Optimization of deep learning and simulation for autonomous vehicles
Burkay Donderici, Burlingame, CA (US)
Assigned to GM Cruise Holdings LLC, San Francisco, CA (US)
Filed by GM Cruise Holdings LLC, San Francisco, CA (US)
Filed on May 31, 2022, as Appl. No. 17/829,027.
Prior Publication US 2023/0382407 A1, Nov. 30, 2023
Int. Cl. B60W 60/00 (2020.01); B60W 50/06 (2006.01); G05B 13/02 (2006.01)
CPC B60W 50/06 (2013.01) [B60W 60/001 (2020.02); G05B 13/027 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method, comprising:
executing a simulated operation of a vehicle in a scene, the vehicle comprising a sensor suite configured to detect the scene and to generate sensor data, the sensor suite comprising one or more sensors;
controlling the operation of the vehicle by using a deep neural network (DNN) that is configured to receive the sensor data as an input and to output a control signal, the vehicle configured to operate in accordance with the control signal;
determining a learning score, the learning score indicating a cost associated with the DNN;
determining a performance score based on an evaluation of a performance of the vehicle during the operation of the vehicle in the scene; and
modifying a parameter associated with the DNN based on the learning score and the performance score.