US 12,260,758 B2
Identifying parkable areas for autonomous vehicles
Chi Yeung Jonathan Ng, San Francisco, CA (US); Anthony Ronald Grue, Thousand Oaks, CA (US); Qichi Yang, Foster City, CA (US); Kevin Poulet, Mountain View, CA (US); Zijian Guo, Sunnyvale, CA (US); and David Harrison Silver, San Carlos, CA (US)
Assigned to Waymo LLC, Mountain View, CA (US)
Filed by Waymo LLC, Mountain Avenue, CA (US)
Filed on Apr. 21, 2021, as Appl. No. 17/236,693.
Prior Publication US 2022/0343763 A1, Oct. 27, 2022
Int. Cl. G08G 1/14 (2006.01); G06N 5/04 (2023.01); G06N 20/00 (2019.01)
CPC G08G 1/143 (2013.01) [G06N 5/04 (2013.01); G06N 20/00 (2019.01)] 21 Claims
OG exemplary drawing
 
1. A method comprising:
identifying, by one or more processors, from logged data, observations of one or more vehicles parked in an area identified as not parkable in a roadgraph;
determining, by the one or more processors based on the observations of the one or more vehicles parked in the area identified as not parkable in the roadgraph whether a sub-portion of an edge in the area identified as not parkable in the roadgraph, corresponds to a parkable area irrespective of the sub-portion of the edge being in the area identified as not parkable in the roadgraph, wherein the edge comprises a distance between two graph nodes of the roadgraph and defines a drivable area in the roadgraph; and
responsive to determining that the sub-portion of the edge in the area identified as not parkable in the roadgraph corresponds to the parkable area, generating, by the one or more processors, map information associated with the area identified as not parkable.