非线性离散时间系统的迭代学习控制

针对一类非线性离散时间系统，在系统参数未知情形下，提出了一种新的开－闭环迭代学习控制律，并给出了控制律的收敛性证明和收敛条件，在新控制律中，学习增益由线性估计模型确定，因此，避免了试凑增益。由机械手实例的仿真结果及收敛性证明，表明了所提出的控制律的有效性，并对初始状态误差和参数估计误差具有良好的鲁棒特性。

ITERATIVE LEARNING CONTROL FOR NONLINEAR DISCRETE TIME SYSTEMS

A new method of iterative learning control is Presented for a class of non linear discretetime systems with unknown parameters.In this method,the learning gain terms are determined by iterative estimation of system dynamics and therefore it is avoided to determine the learning gain terms through the objectivity.Proof and conditon are stated for convergence of the proposed method.Simulation study for a robot manipulator has been carried out.The results of simulation show the validity of the puoposed method,its robustness to the initial state error and the estimating error for system dynamics.