US 11,814,073 B2
Learning based controller for autonomous driving
Shu Jiang, Sunnyvale, CA (US); Qi Luo, Sunnyvale, CA (US); Jinghao Miao, Sunnyvale, CA (US); Jiangtao Hu, Sunnyvale, CA (US); Yu Wang, Sunnyvale, CA (US); Jinyun Zhou, Sunnyvale, CA (US); Jiaming Tao, Sunnyvale, CA (US); and Kecheng Xu, Sunnyvale, CA (US)
Assigned to BAIDU USA LLC, Sunnyvale, CA (US)
Filed by Baidu USA LLC, Sunnyvale, CA (US)
Filed on Mar. 18, 2020, as Appl. No. 16/823,141.
Prior Publication US 2021/0291862 A1, Sep. 23, 2021
Int. Cl. B60W 60/00 (2020.01)
CPC B60W 60/0011 (2020.02) [B60W 60/0017 (2020.02); B60W 60/00182 (2020.02)] 20 Claims
OG exemplary drawing
 
1. A method for operating an autonomous driving vehicle (ADV), the method comprising:
generating a control command with an MPC (model predictive control) controller, the MPC controller including an optimizer using a cost function configured with one or more weights associated with cost terms of the cost function wherein the control command is associated with a first trajectory of the ADV;
performing a simulation in a three-dimensional virtual environment with a dynamic model of the ADV that includes one or more physical attributes that are representative of a current state of the ADV including a passenger or cargo in the ADV;
in the simulation, applying the control command generated by the MPC to the dynamic model of the ADV to simulate a behavior of the ADV with the passenger or the cargo in the ADV and generate a second trajectory of the dynamic model in the three-dimensional virtual environment;
adjusting the one or more weights based on an evaluation of the dynamic model with the passenger or the cargo in response to the control command of the MPC controller in the simulation to adjust the second trajectory of the dynamic model to track the first trajectory of the ADV, resulting in an adjusted cost function of the MPC controller wherein adjusting the one or more weights includes decreasing one of the one or more weights that is associated with a speed cost term, to increase speed more aggressively, in response to one or more current environmental conditions indicating low vehicle traffic;
generating, with the MPC controller having the adjusted cost function, a second control command used to effect movement of the ADV; and
controlling movement of the ADV according to the second control command.