CPC G06T 7/0004 (2013.01) [A01B 3/26 (2013.01); A01B 79/005 (2013.01); G06F 18/214 (2023.01); G06F 18/24 (2023.01); G06K 7/10366 (2013.01); G06K 19/0723 (2013.01); G06N 20/00 (2019.01); G06T 7/593 (2017.01); H04N 13/204 (2018.05); G06T 2207/20081 (2013.01); G06T 2207/30164 (2013.01); G06T 2207/30204 (2013.01)] | 20 Claims |
1. A method of detecting tilling shank malfunctions by an autonomous farming vehicle, comprising:
accessing, in real-time, image data corresponding to and representative of a state of each tilling shank in a set of tilling shanks as the autonomous farming vehicle pulls the set of tilling shanks through earth while performing an autonomous routine, the image data capturing light reflected off each tilling shank in the set of tilling shanks, and the image data captured by a camera of the autonomous farming vehicle;
applying a machine-learned model to the image data, the machine-learned model trained on training image data representative of operational tilling shanks and malfunctioning tilling shanks, the machine-learned model configured to classify, for each tilling shank in the set of tilling shanks, whether the tilling shank is operational or malfunctioning based on a shape of the light reflected off the tilling shank; and
in response to a tilling shank in the set of tilling shanks being classified as malfunctioning, modifying the autonomous routine of the autonomous farming vehicle.
|