US 12,241,757 B1
Generation of weighted map data for autonomous vehicle localization
Greg Coombe, Mountain View, CA (US); and Yekeun Jeong, Sunnyvale, CA (US)
Assigned to AURORA OPERATIONS, INC, Mountain View, CA (US)
Filed by Aurora Operations, Inc., Mountain View, CA (US)
Filed on Jun. 27, 2022, as Appl. No. 17/850,340.
Application 17/850,340 is a continuation of application No. 17/849,049, filed on Jun. 24, 2022.
Int. Cl. G01C 21/00 (2006.01); B60W 60/00 (2020.01)
CPC G01C 21/3867 (2020.08) [B60W 60/001 (2020.02); B60W 2556/40 (2020.02)] 18 Claims
OG exemplary drawing
 
1. A computer-implemented method for generating a weighted map used for autonomous vehicle localization, comprising:
accessing stored map data describing the environment, wherein the map data includes a plurality of map data units that describe a plurality of geometric elements for a plurality of objects in the environment;
generating a weight associated with a respective map data unit of the plurality of map data units, wherein the weight is indicative of a relative stability of a respective geometric element that is described by the respective map data unit, wherein generating the weight includes performing geometric analysis of an orientation and/or position of the respective geometric element described by the respective map data unit in the environment; and
storing and communicating the generated weight to an autonomous vehicle control system of an autonomous vehicle to cause the autonomous vehicle control system to use the generated weight to weight a contribution of the respective map data unit in an alignment operation that determines a pose for the autonomous vehicle by aligning a plurality of sensor returns obtained from one or more sensors of the autonomous vehicle with the plurality of map data units and to control a trajectory of the autonomous vehicle using the determined pose.