	US 12,235,651 B1
	Autonomous navigation correction system
Lionel Gueguen, Erie, CO (US); Ben James Kadlec, Boulder, CO (US); Denver H. Dash, Longmont, CO (US); and Drew Steedly, Woodinville, WA (US)
Assigned to Amazon Technologies, Inc., Seattle, WA (US)
Filed by Amazon Technologies, Inc., Seattle, WA (US)
Filed on Mar. 29, 2021, as Appl. No. 17/215,594.
Int. Cl. G05D 11/00 (2006.01); G05D 1/00 (2006.01); G06N 20/00 (2019.01); G06T 7/50 (2017.01); G06T 7/90 (2017.01)

CPC G05D 1/0238 (2013.01) [G05D 1/0274 (2013.01); G06N 20/00 (2019.01); G06T 7/50 (2017.01); G06T 7/90 (2017.01); G06T 2207/10028 (2013.01); G06T 2207/20081 (2013.01)]

16 Claims

1. A method comprising:

determining first data corresponding to color information associated with an environment;

determining second data corresponding to first depth information associated with the environment;

determining second depth information based on the color information associated with the first data;

determining third data based on the second depth information, the third data indicative of a three-dimensional representation of the first data;

determining fourth data based on the second data, the fourth data indicative of a point cloud corresponding to the second data;

determining fifth data by combining the fourth data with the third data;

determine a first obstacle based on the fifth data; and

determine the first obstacle is traversable based on the fifth data.