US 11,655,028 B2
Reinforcement learning based system for aerial imagery acquisition using drone following target vehicle
Mohammed H. Al Qizwini, Macomb, MI (US); and David H. Clifford, Royal Oak, MI (US)
Assigned to General Motors LLC, Detroit, MI (US)
Filed by General Motors LLC, Detroit, MI (US)
Filed on Feb. 19, 2020, as Appl. No. 16/794,407.
Prior Publication US 2021/0256255 A1, Aug. 19, 2021
Int. Cl. G05D 1/12 (2006.01); G06N 3/08 (2023.01); G06V 20/10 (2022.01); G05D 1/00 (2006.01); G01S 19/43 (2010.01); B64C 39/02 (2023.01); G06T 7/73 (2017.01); B64U 101/30 (2023.01)
CPC G06V 20/182 (2022.01) [B64C 39/024 (2013.01); G01S 19/43 (2013.01); G05D 1/0094 (2013.01); G05D 1/12 (2013.01); G06N 3/08 (2013.01); G06T 7/74 (2017.01); B64U 2101/30 (2023.01); B64U 2201/104 (2023.01)] 20 Claims
OG exemplary drawing
 
1. A method of surveying roads comprising:
generating a dynamic flight plan for a drone using a vehicle traveling on a road as a target, wherein the dynamic flight plan includes instructions for movement of the drone;
controlling the drone as a function of position of the vehicle based on the dynamic flight plan, wherein the controlling the drone as a function of position of the vehicle includes using reinforced learning;
maintaining, based on the controlling, line of sight with the drone while the drone with an onboard camera follows the vehicle and captures images of the road being traveled by the vehicle using the onboard camera; and
using the reinforced learning, determining a position of the drone relative to the vehicle at a granular level and providing a control policy that the drone be within a particular relative position from the vehicle and that the drone maintains the position relative to the vehicle at any given epoch of movement for subsequently aligning aerial imagery captured by the drone with ground imagery.