US 12,077,177 B2
Autonomous vehicle control and map accuracy determination based on predicted and actual trajectories of surrounding objects and vehicles
Sachin Hagaribommanahalli, Sunnyvale, CA (US); Christopher Ostafew, Mountain View, CA (US); and David Ilstrup, Santa Cruz, CA (US)
Assigned to Nissan North America, Inc., Franklin, TN (US)
Filed by Nissan North America, Inc., Franklin, TN (US)
Filed on May 26, 2021, as Appl. No. 17/330,839.
Prior Publication US 2022/0379910 A1, Dec. 1, 2022
Int. Cl. B60W 40/09 (2012.01); B60W 60/00 (2020.01); G06K 9/00 (2022.01); G06V 20/59 (2022.01)
CPC B60W 60/001 (2020.02) [B60W 40/09 (2013.01); G06V 20/597 (2022.01); B60W 2540/215 (2020.02); B60W 2540/22 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method, comprising:
detecting a road user;
determining respective predicted data for the road user;
storing, in a data structure, the respective predicted data;
storing, in the data structure, actual data of the road user;
storing, in the data structure, map data corresponding to the actual data, wherein the map data is obtained from a high-definition map;
obtaining an average prediction displacement error using at least one of the actual data and at least two corresponding respective predicted data, wherein the average prediction displacement error is a numerical value indicating a distance displacement that an actual trajectory of the road user exhibits compared to a predicted trajectory of the road user through time, and wherein the at least two corresponding respective predicted data are predictions of the at least one of the actual data;
categorizing prediction accuracy based on the average prediction displacement error;
obtaining an average map displacement error based on a comparison of at least one of the actual data and at least two corresponding respective map data, wherein the average map displacement error indicates distance differences between a mapped driveline of a lane and how the road user actually traverses the lane;
categorizing map accuracy based on the average map displacement error; and
determining a scene understanding based on the categorized prediction accuracy and the categorized map accuracy.