US 11,814,076 B2
System and method for autonomous vehicle performance grading based on human reasoning
Vincent De Sapio, Westlake Village, CA (US); Steven W. Skorheim, Los Angeles, CA (US); and Iman Zadeh, Los Angeles, CA (US)
Assigned to GM GLOBAL TECHNOLOGY OPERATIONS LLC, Detroit, MI (US)
Filed by GM GLOBAL TECHNOLOGY OPERATIONS LLC, Detroit, MI (US)
Filed on Dec. 3, 2020, as Appl. No. 17/110,750.
Prior Publication US 2022/0176993 A1, Jun. 9, 2022
Int. Cl. B60W 60/00 (2020.01); B60W 50/06 (2006.01); G06N 3/004 (2023.01)
CPC B60W 60/0011 (2020.02) [B60W 50/06 (2013.01); G06N 3/004 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for operating an autonomous vehicle, comprising:
inputting a training set of data to a cognitive system;
operating the cognitive system to generate a planned action for operating the autonomous vehicle in response to the training set of data;
evaluating the planned action to generate a system performance grade based on at least one basis metric weighted by a coefficient, the at least one basis metric including a deviation from a safe following distance;
showing the training set of data to one or more humans to obtain a human-based performance grade;
updating the cognitive system by adjusting the coefficient of the at least one basis metric based on a comparison of the system performance grade to the human-based performance grade; and
controlling autonomous driving of the autonomous vehicle using the cognitive system.