US 12,145,628 B1
Driving difficulty maps for autonomous vehicles
Xiaoyue Zhao, Saratoga, CA (US); Henning Hohnhold, San Francisco, CA (US); Yin-Hsiu Chen, San Francisco, CA (US); Xiang Gao, Mountain View, CA (US); Ajay Joshi, Mountain View, CA (US); and Kevin Cao, Mountain View, CA (US)
Assigned to Waymo LLC, Mountain View, CA (US)
Filed by WAYMO LLC, Mountain View, CA (US)
Filed on May 9, 2023, as Appl. No. 18/195,094.
Application 18/195,094 is a continuation of application No. 17/347,685, filed on Jun. 15, 2021, granted, now 11,685,408.
This patent is subject to a terminal disclaimer.
Int. Cl. G01C 5/02 (2006.01); B60W 60/00 (2020.01); G01C 21/00 (2006.01); G07C 5/02 (2006.01)
CPC B60W 60/005 (2020.02) [G01C 21/3841 (2020.08); G07C 5/02 (2013.01); B60W 2556/40 (2020.02); B60W 2556/45 (2020.02)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
using, by one or more processors, first log data generated by a vehicle driving in a first geographic area to train a machine learned model using the first log data;
inputting, by the one or more processors, second log data generated by a vehicle driving in a second geographic area into the machine learned model in order to generate an output identifying a likelihood of a vehicle driving in an autonomous driving mode requiring a disengage from the autonomous driving mode in the second geographic area is different from the first geographic area;
generating, by the one or more processors, a driving difficulty map for the second geographic area based on the output; and
using the driving difficulty map to select a route and control a vehicle in an autonomous driving mode using the route.