基于改进型BP神经网络的四旋翼控制系统  

Four-rotor Control System Based on Improved BP Neural Network

作　　者：余后明 刘彦臣[1] 郑士振 常建龙 Yu Houming;Liu Yanchen;Zheng Shizhen;Chang Jianlong(College of Mechanical and Electrical Engineering,The North University of China,Taiyuan 030051,China)

机构地区：[1]中北大学机电工程学院,山西太原030051

出　　处：《甘肃科学学报》2019年第2期87-91,共5页Journal of Gansu Sciences

基　　金：山西省青年自然科学基金(201701D221016);山西省应用基础研究计划项目(201701D221146)

摘　　要：为了实现对四旋翼无人机的自稳定控制,首先对四旋翼无人机进行了动力学建模,提出了一种改变学习率的BP神经网络算法与PID控制相结合的姿态控制方法,并在相同环境下与常规PID控制器进行了仿真试验对比。仿真试验结果表明:基于改进型BP神经网络的PID控制器能够有效地实现无人机的自稳定控制,相比于常规PID控制器,基于改进型BP神经网络的PID控制器具有响应速度快\,超调量低\,鲁棒性强等优点。In order to realize the self-stability control of the four-rotor UAV, the dynamics modeling of the four-rotor UAV is firstly built, and an attitude control method combining BP neural network algorithm with learning rate and PID control is proposed. The simulation test was compared with the conventional PID controller under the same environment. The simulation results show that the PID controller based on the improved BP neural network can effectively realize the self-stability control of the UAV. Compared with the conventional PID controller, the PID controller based on the improved BP neural network has a good feature of fast response speed, low overshoot and strong robustness.

关 键 词：四旋翼无人机 动力学建模 学习率 BP神经网络 

分 类 号：V279[航空宇航科学与技术—飞行器设计]