US 11,861,896 B1
Autonomous aerial navigation in low-light and no-light conditions
Samuel Shenghung Wang, Mountain View, CA (US); Vladimir Nekrasov, Adelaide (AU); Ryan David Kennedy, Redwood City, CA (US); Gareth Benoit Cross, Mountain View, CA (US); Peter Benjamin Henry, San Francisco, CA (US); Kristen Marie Holtz, Menlo Park, CA (US); Hayk Martirosyan, San Francisco, CA (US); Abraham Galton Bachrach, Redwood City, CA (US); and Adam Parker Bry, Redwood City, CA (US)
Assigned to Skydio, Inc., San Mateo, CA (US)
Filed by Skydio, Inc., Redwood City, CA (US)
Filed on Mar. 29, 2022, as Appl. No. 17/707,841.
Claims priority of provisional application 63/168,827, filed on Mar. 31, 2021.
Int. Cl. G06V 20/17 (2022.01); G06V 10/82 (2022.01); G06V 10/30 (2022.01); H04N 5/33 (2023.01); G06T 5/00 (2006.01); G06T 3/40 (2006.01); G05D 1/10 (2006.01); B64C 39/02 (2023.01); G06V 10/60 (2022.01); B64U 101/30 (2023.01)
CPC G06V 20/17 (2022.01) [B64C 39/024 (2013.01); G05D 1/101 (2013.01); G06T 3/4038 (2013.01); G06T 5/008 (2013.01); G06V 10/30 (2022.01); G06V 10/60 (2022.01); G06V 10/82 (2022.01); H04N 5/33 (2013.01); B64U 2101/30 (2023.01); B64U 2201/10 (2023.01); G06T 2207/10024 (2013.01); G06T 2207/10032 (2013.01); G06T 2207/20024 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/20182 (2013.01); G06T 2207/30252 (2013.01)] 20 Claims
OG exemplary drawing
 
1. An apparatus, comprising:
a memory configured to store instructions for training a learning model for use with unmanned aerial vehicle navigation; and
a processor configured to execute the instructions stored in the memory to:
produce a first image including infrared data from an infrared light onboard an unmanned aerial vehicle by simulating a reflection of the infrared data to determine a simulated infrared illumination range within an environment depicted by a first copy of input image data;
perform range-based darkening against a second copy of the input image data to produce a second image including darkened RGB color data;
combine the first image and the second image to produce a combined image; and
train the learning model based on the combined image.