摘要
为了降低无人机轨迹跟踪误差,提高系统抗干扰能力,对反步(Backstepping)法进行改进提出一种基于反步神经网络(BSP-ANN)的无人机轨迹跟踪方法。首先,建立了四旋翼无人机运动学模型;然后,结合Backstepping方法在无人机的姿态控制、轨迹跟踪控制系统中引入Sigma-Pi神经网络,同时设计Sigma-Pi神经网络控制率,并证明该控制率满足Lyapunov意义下的系统稳定;最后,分别给出了相应的仿真实验。仿真结果表明:该算法可以有效降低跟踪误差,缩短无人机跟踪时间,同时可以提高系统的抗干扰能力。
To reduce the trajectory tracking error of the UAV and improve the anti-jamming ability of the system,a new trajectory tracking algorithm of the UAV is proposed based on BSP-ANN.A dynamic model for the Quadrotor UAV is given.Based on the Backstepping method,the Sigma-Pi ANN is introduced into the position control system and attitude control system of the Quadrotor UAV.The Sigma-Pi ANN control law is designed,and proving the systems stability in the sense of Lyapunov function.The corresponding simulations are performed using MATLAB.Simulation results show that with the BSP-ANN method,the trajectory tracking performance of the UAV can be improved by reducing the trajectory tracking error,decreasing the tracking time,and improving the anti-interference ability of the system.
作者
陈志明
牛康
李磊
吴云华
华冰
CHEN Zhiming;NIU Kang;LI Lei;WU Yunhua;HUA Bing(Micro Satellite Research Center, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China)
出处
《航空学报》
EI
CAS
CSCD
北大核心
2018年第6期177-184,共8页
Acta Aeronautica et Astronautica Sinica
基金
国家自然科学基金(61673212)
航空科学基金(20150852013)
江苏省自然科学基金(BK20161490)
上海市优秀学科带头人计划(14XD1423300)~~
