US 12,270,726 B2
Hyperplane search-based vehicle test
James Hryniw, San Francisco, CA (US)
Assigned to GM Cruise Holdings LLC, San Francisco, CA (US)
Filed by GM Cruise Holdings LLC, San Francisco, CA (US)
Filed on Aug. 4, 2022, as Appl. No. 17/881,520.
Prior Publication US 2024/0044745 A1, Feb. 8, 2024
Int. Cl. G06N 20/10 (2019.01); B60W 60/00 (2020.01); G01M 17/007 (2006.01)
CPC G01M 17/007 (2013.01) [B60W 60/00 (2020.02); G06N 20/10 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented system, comprising:
one or more non-transitory computer-readable media storing instructions, when executed by one or more processing units, cause the one or more processing units to perform operations comprising:
generating a first plurality of vehicle test scenarios, each corresponding to one of a first plurality of sampling points in an N-dimensional parameter space associated with a vehicle capability, wherein N is a positive integer;
executing a vehicle compute process in each vehicle test scenario of the first plurality of vehicle test scenarios to generate a first test result for each respective vehicle test scenario, wherein executing the vehicle compute process in each vehicle test scenario of the first plurality of vehicle test scenarios includes performing perception, prediction, planning, and control algorithms of an autonomous vehicle in a simulated environment based on each vehicle test scenario of the first plurality of vehicle test scenarios to generate the first test result for each respective vehicle test scenario;
computing, based on the first test results, a first hyperplane in the N-dimensional parameter space; and
generating, based on the first hyperplane, a second plurality of vehicle test scenarios, each corresponding to one of a second plurality of sampling points in the N-dimensional parameter space,
wherein generating the first plurality of vehicle test scenarios, executing the vehicle compute process for the first plurality of vehicle test scenarios, and generating the second plurality of vehicle test scenarios based on the first hyperplane tests and validates algorithms, machine learning models, and neural networks of the vehicle compute process of the autonomous vehicle for managing operation of the autonomous vehicle in a real-world environment.