基于附加惯性项BP神经网络的四旋翼无人机姿态控制研究  

Study of Applying BP Neural Network with Inertia Term Self-tuning to Attitude Stability of Quadrotor Unmanned Aerial Vehicle

作　　者：钟海鑫 丘森辉[1] 罗晓曙[1] 唐堂[1] 杨力[1] 赵帅[1] 

机构地区：[1]广西师范大学电子工程学院,广西桂林541004

出　　处：《广西师范大学学报（自然科学版）》2017年第2期24-31,共8页Journal of Guangxi Normal University:Natural Science Edition

基　　金：国家自然科学基金(11262004);广西多源信息挖掘与安全重点实验室开放基金(MIMS15-06);广西信息科学实验中心基金(KA1430);广西研究生教育创新计划项目(XYCSZ2017051)

摘　　要：针对传统PID控制不能实时更新整定的K_p、K_i、K_d参数,以及控制精度不高等缺点,在结合四旋翼无人机自身特性的基础上,本文提出一种附加惯性项的BP神经网络与PID控制结合的姿态控制方法,并对惯性系数进行修改,来实现无人机在受干扰情况下飞行过程的姿态控制。仿真实验分析表明:与BP神经网络参数自整定PID控制和传统PID控制相比,该控制方法提高了系统的抗扰性、鲁棒性和动态性能,从而对于提高无人机的姿态控制具有较好的实际参考价值。Due parameters in to the shortcomings that traditional PID control fails to update the tuning of Kp, Ki, Kd real time and the control accuracy is not high, this paper puts forward a kind of attitude stability control method of the combination between BP neural network with inertia term and PID control. And the inertia coefficient is modified to achieve Quadrotor Unmanned Aerial Vehicle（QUAV）＇ s attitude control during the flight under disturbance. The simulation experiment analysis presents that when compared with the BP neural network parameters self-tuning PID control and the traditional PID control, this control method improves the performance of anti-disturbance and robustness and the dynamic performance, thus holding practical value for reference to improve the attitude stability control of QUAV.

关 键 词：PID控制 BP神经网络 参数自整定 抗扰性 鲁棒性 

分 类 号：TP273.2[自动化与计算机技术—检测技术与自动化装置] TP271[自动化与计算机技术—控制科学与工程]