基于间接型迭代学习控制的四旋翼轨迹跟踪  

Quadrotor trajectory tracking based on indirect iterative learning control

作　　者：焦金彦 薛建平[1] 董新民[1] 刘娇龙[1] 

机构地区：[1]空军工程大学航空航天工程学院,陕西西安710038

出　　处：《现代电子技术》2017年第23期113-118,共6页Modern Electronics Technique

摘　　要：针对反推控制方法在四旋翼飞行器轨迹跟踪中存在动态误差的问题,提出改善轨迹跟踪性能的间接迭代学习控制算法。首先将飞行器分为姿态和高度全驱动控制通道和水平轨迹欠驱动控制通道;然后基于反推控制方法分别设计了全驱动和欠驱动通道的稳定控制器,进而采用PID型迭代学习算法对反推控制器参考输入信号进行更新优化;最后,采用压缩映射原理和λ范数理论证明了算法的收敛性。仿真实验结果表明,PID型迭代学习律能够有效改善反推控制器的控制效果,实现对期望轨迹的完全跟踪。The backstepping control method has dynamic error in the trajectory tracking of the quadrotors,so an indirect iterative learning control(ILC)algorithm is proposed to improve the performance of trajectory tracking. The control of the aircraft involves the attitude and height full-drive control channel,and horizontal trajectory underdrive control channel. A stable controller of the full-drive and underdrive control channels was designed on the basis of backstepping control method. The PID iterative learning algorithm is used to perform the update optimization for the reference input signal of the backstepping controller. The contracting mapping principle and λ-norm theory are adopted to verify the convergence of the algorithm. The simulation results show that the PID iterative learning law can improve the control effect of the backstepping controller effectively,and realize the full tracking for the expected trajectory.

关 键 词：四旋翼飞行器 反推控制 迭代学习控制 姿态控制 轨迹跟踪 

分 类 号：TN876-34[电子电信—信息与通信工程] TP273[自动化与计算机技术—检测技术与自动化装置]