US 12,221,764 B2
Automated ground leveling using artificial neural network
Tien-Chuong Lim, Grovetown, GA (US); Christopher T. McCord, Thomson, GA (US); and Benjamin Seth Peavy, Harlem, GA (US)
Assigned to DEERE & COMPANY, Moline, IL (US)
Filed by Deere & Company, Moline, IL (US)
Filed on Aug. 17, 2021, as Appl. No. 17/445,228.
Claims priority of provisional application 62/706,940, filed on Sep. 18, 2020.
Prior Publication US 2022/0090352 A1, Mar. 24, 2022
Int. Cl. E02F 3/84 (2006.01); A01B 63/111 (2006.01); G06N 3/048 (2023.01); E02F 3/76 (2006.01)
CPC E02F 3/844 (2013.01) [A01B 63/111 (2013.01); G06N 3/048 (2023.01); E02F 3/7631 (2013.01)] 24 Claims
OG exemplary drawing
 
1. A method of grading a ground surface with a traction unit having a three-point hitch system, and a scraper implement attached to the three-point hitch system and drawn behind the traction unit, the method comprising:
sensing a pitch angle of the traction unit at a current location of the traction unit with an angle sensor, and communicating the pitch angle of the traction unit to a controller;
calculating a leveling variation with the controller, wherein the leveling variation is a numerical difference between a current elevation of a cutting edge of the scraper implement and a desired elevation of the ground surface at a current location of the cutting edge of the scraper implement; and
generating a hitch control signal with the controller based on the pitch angle of the traction unit and the leveling variation, wherein the hitch control signal is operable to control the three-point hitch system to adjust a vertical position of the cutting edge of the scraper implement relative to the traction unit.