Research on Optimal Control Method of Four Rotor UAV Based on BP Neural Network

Abstract

With the development of related technical fields, the application scenarios of four rotor unmanned aerial vehicle (UAV) is becoming wider and wider. Especially in the field of power inspection, UAV inspection has gradually replaced manual inspection, forming a new working mode. At present, the unmanned aerial vehicle inspection technology applied to transmission lines has become increasingly mature. However, UAV technology has only been gradually applied to the patrol inspection of overhead power distribution network in recent years. The traditional proportional integral derivative (PID) control method of unmanned aerial vehicle (UAV) is difficult to meet the needs of UAV patrol inspection work in terms of control accuracy and response speed. To solve this problem, this paper uses back propagation neural net to optimize the traditional control method. Appropriate control parameters are trained by online learning. The improved control core unit has the function of automatic setting of control parameters. This enables the UAV to adapt to the changing flight environment and fly more smoothly. Finally, the improved back propagation neural net PID controller is used to simulate the system model. The research results have a positive role in promoting the development of unmanned aerial vehicle inspection technology for distribution lines.