基于FHM的四旋翼飞行器悬停控制方法  

Hovering Control Method for Four Rotor Aircraft Based on FHM

作　　者：陈侠 梁红利 尹立远 CHEN Xia;LIANG Hong-li;YIN Li-yuan(School of Automation,Shenyang Aerospace University,Shen Yang 110136,China)

机构地区：[1]沈阳航空航天大学自动化学院

出　　处：《计算机仿真》2019年第9期38-46,74,共10页Computer Simulation

基　　金：国家自然科学基金项目(61074159);航空科学基金(2016ZC54011);辽宁省自然科学基金(2015020063)

摘　　要：针对四旋翼飞行器运动的建模与悬停控制问题,提出一种基于模糊双曲正切模型(Fuzzy hyperbolic tangent model,FHM)的四旋翼飞行器悬停控制方法。首先,分别建立四旋翼飞行器运动的模糊双曲正切模型和神经网络模型。其次,根据反向传播(Back Propagation,BP)算法辨识神经网络模型参数,得到参数调解率公式。然后,在模糊双曲正切模型基础上设计一种最优控制器,得到四旋翼飞行器运动系统的控制率。最后通过数值仿真证明,模糊双曲正切模型精确描述了四旋翼飞行器的运动过程,很好地反映了运动规律。上述方法使系统控制偏差的稳定值小于0.1%,实现四旋翼飞行器的悬停控制,并且与模糊自整定PID方法相比,位置偏差的稳定值更小,系统的收敛速度更快。Aim at the modeling and hovering control of four rotor aircraft,a hovering control method based on fuzzy hyperbolic tangent model is proposed.Firstly,the fuzzy hyperbolic tangent model and neural network model of four rotor aircraft were established respectively.Secondly,the Back Propagation(BP)algorithm was used to identify the parameters of the neural network model,and the formula of the weight coefficient mediation rate was derived,as well as the parameters of the model were obtained.Then,based on the fuzzy hyperbolic tangent model,an optimal controller was designed to obtain the control rate of the four rotor aircraft's motion system.Finally,numerical simulation shows that the fuzzy hyperbolic tangent model describes the movement process of the four rotor aircraft accurately,which reflects the movement law very well.The control deviation is less than 0.1%,and the hovering control of the four rotor aircraft is realized.Compared with the fuzzy self-tuning PID method,the control error of the system is smaller and the convergence speed is faster.

关 键 词：四旋翼飞行器 神经网络 最优控制 

分 类 号：V19[航空宇航科学与技术—人机与环境工程]