US 11,915,440 B2
Generation of structured map data from vehicle sensors and camera arrays
Nikhil Naikal, San Francisco, CA (US); Alonso Patron-Perez, Elkins Park, PA (US); Alexander Marques, San Jose, CA (US); John Kua, San Francisco, CA (US); Aaron Matthew Bestick, San Francisco, CA (US); Christopher D. Thompson, Alexandria, VA (US); and Andrei Claudiu Cosma, Emeryville, CA (US)
Assigned to Velodyne Lidar USA, Inc., San Jose, CA (US)
Filed by VELODYNE LIDAR USA, INC., San Jose, CA (US)
Filed on Apr. 26, 2021, as Appl. No. 17/240,307.
Application 17/240,307 is a continuation of application No. 16/254,508, filed on Jan. 22, 2019, granted, now 11,004,224.
Prior Publication US 2021/0248768 A1, Aug. 12, 2021
This patent is subject to a terminal disclaimer.
Int. Cl. G06T 7/579 (2017.01); G06T 7/73 (2017.01)
CPC G06T 7/579 (2017.01) [G06T 7/74 (2017.01); G06T 2207/10016 (2013.01)] 21 Claims
OG exemplary drawing
 
1. A method of generating structured map data, the method comprising:
receiving time-stamped sensor data from at least one sensor disposed on a vehicle;
receiving video data comprising a plurality of video streams provided by a respective plurality of cameras disposed on the vehicle, wherein the video data are time-stamped based on a timing signal provided by a global positioning system (GPS):
determining, based on the time-stamped sensor data, a plurality of spatial values each comprising a position of the vehicle and an associated time stamp, the plurality of spatial values including a particular spatial value indicating a particular position of the vehicle and a particular associated time stamp; and
adjusting the particular spatial value indicating the particular position of the vehicle based on two or more parallax differences in two or more respective relative positions of a fixed feature depicted in a plurality of image records captured in the video data.