齿隙非线性输入系统的迭代学习控制

针对一类具有输入齿隙特性的非线性系统, 提出一种实现有限作业区间轨迹跟踪的迭代学习控制方法. 在系统不确定项可参数化的情形下, 基于类Lyapunov方法设计迭代学习控制器, 回避了常规迭代学习控制中受控系统非线性特性需满足全局Lipschitz连续条件的要求. 对未知时变参数进行泰勒级数展开, 参数估计采用微分学习律, 并在控制器设计中, 采用双曲函数处理级数展开后的余项以及齿隙特性里的有界误差项, 以保证控制器可导, 且可抑制颤振. 引入一级数收敛序列确保系统输出完全跟踪期望轨迹, 且闭环系统所有信号有界.

Iterative Learning Control of a Class of Nonlinear Systems with Input Backlash

In this paper, an iterative learning control method is presented for a class of time-varying nonlinear systems with input backlash. In the circumstance of parameterized nonlinear uncertainties, unknown time-varying parameters are expanded into Taylor series and differential learning mechanisms are used to handle non-global Lipschitz nonlinearities in system dynamics. In the controller design, hyperbolic tangent function is introduced to eliminate the influence of the remaining term and the bounded error term due to input backlash, and simultaneously to guarantee the differentiability of the controller and suppress the chattering. Through Lyapunov-like synthesis and with the aid of a convergent series, the complete tracking over a pre-specified time interval is achieved and boundedness of all the signals in the closed-loop system is ensured.