US 11,815,891 B2
End dynamics and constraints relaxation algorithm on optimizing an open space trajectory
Runxin He, Sunnyvale, CA (US); Yu Wang, Sunnyvale, CA (US); Jinyun Zhou, Sunnyvale, CA (US); Qi Luo, Sunnyvale, CA (US); Jinghao Miao, Sunnyvale, CA (US); Jiangtao Hu, Sunnyvale, CA (US); Jingao Wang, Sunnyvale, CA (US); Jiaxuan Xu, Sunnyvale, CA (US); and Shu Jiang, Sunnyvale, CA (US)
Assigned to BAIDU USA LLC, Sunnyvale, CA (US)
Filed by Baidu USA LLC, Sunnyvale, CA (US)
Filed on Oct. 22, 2019, as Appl. No. 16/659,963.
Prior Publication US 2021/0116916 A1, Apr. 22, 2021
Int. Cl. G05D 1/00 (2006.01); G05D 1/02 (2020.01)
CPC G05D 1/0088 (2013.01) [G05D 1/0214 (2013.01); G05D 1/0221 (2013.01); G05D 2201/0213 (2013.01)] 21 Claims
OG exemplary drawing
 
1. A computer-implemented method for operating an autonomous driving vehicle, the method comprising:
determining a target function for an open space model based on two or more obstacles and map information within a proximity of an autonomous driving vehicle (ADV);
iteratively, until a predetermined converged condition is satisfied,
determining a first trajectory that initializes a first set of variables including dual variables that indicate a distance between each of the two or more obstacles and the ADV,
performing a first quadratic programming (QP) optimization on the target function based on the first trajectory while fixing the first set of variables of the target function, and
performing a second QP optimization on the target function based on a result of the first QP optimization while fixing a second set of variables of the target function, wherein second QP optimization includes maximizing a sum of distances from the ADV to a boundary of each obstacle in a set of two or more obstacles, over each of a plurality of points in time of the first trajectory, and wherein the target function includes a penalty for deviation based on the second QP optimization on the target function;
generating a second trajectory based on results of the first and the second QP optimizations; and
controlling the ADV autonomously according to the second trajectory.