US 12,405,611 B2
Autonomous machine navigation in lowlight conditions
Michael Jason Ramsay, Verona, PA (US); David Arthur LaRose, Rochester Hills, MI (US); Zachary Irvin Parker, Pittsburgh, PA (US); Matthew John Alvarado, Mars, PA (US); Stephen Paul Elizondo Landers, Pittsburgh, PA (US); and David Ian Robinson, Hawke's Bay (NZ)
Assigned to THE TORO COMPANY, Bloomington, MN (US)
Appl. No. 17/439,465
Filed by THE TORO COMPANY, Bloomington, MN (US)
PCT Filed Apr. 9, 2020, PCT No. PCT/US2020/027471
§ 371(c)(1), (2) Date Sep. 15, 2021,
PCT Pub. No. WO2020/210493, PCT Pub. Date Oct. 15, 2020.
Claims priority of provisional application 62/831,347, filed on Apr. 9, 2019.
Prior Publication US 2022/0151144 A1, May 19, 2022
Int. Cl. G05D 1/00 (2024.01); A01D 34/00 (2006.01); A01D 101/00 (2006.01); G06T 7/80 (2017.01); G06V 10/42 (2022.01); G06V 20/10 (2022.01); G06V 20/56 (2022.01); H04N 5/58 (2006.01); H04N 23/73 (2023.01)
CPC G05D 1/0246 (2013.01) [A01D 34/008 (2013.01); G06T 7/80 (2017.01); G06V 10/42 (2022.01); G06V 20/10 (2022.01); G06V 20/56 (2022.01); H04N 5/58 (2013.01); H04N 23/73 (2023.01); A01D 2101/00 (2013.01)] 21 Claims
OG exemplary drawing
 
1. A method for automatic camera capture configuration of a vision and navigation system of an autonomous machine comprising:
capturing image data of a scene using one or more cameras of the autonomous machine configured with at least one initial camera capture parameter that comprises one or more of exposure time, gain, and active lighting intensity;
comparing pixel intensities of the image data to a pixel intensity threshold to determine if the image data is well exposed;
generating irradiance image data based on applying a calibrated irradiance map to estimate scene brightness of the image data;
generating simulated image data based on an estimate of the image data that would be captured in view of the estimated scene brightness if a test camera capture parameter was used to capture the image data instead of initial camera capture parameter;
determining a simulated pixel intensity from the simulated image data; and
generating at least one updated camera capture parameter based on the test camera capture parameter in response to determining that the simulated pixel intensity is well exposed.