Abstract: An algorithm of model reference iterative learning control (MR-ILC) was proposed for nonlinear systems with periodic and non-periodic uncertain disturbances. The objective is to make the state variables of a controlled system can be steered to follow the state variables of a reference model that does not necessarily have the same structure as the controlled system on condition that the controlled system is satisfied with assumptions and adopted with matching learning rate. The λ-norm was adopted as the topological measure in proof of the convergence of the system. The effectiveness of proposed scheme was illustrated by simulation results.

Key words: 　Non-linearity , Model reference , Iterative learning control , λ-norm

摘要： 　    针对存在周期或非周期不确定干扰的非线性系统，提出一种模型参考迭代学习控制算法（ＭＲＩＬＣ），其目标是在被控系统满足假设条件并采用相应学习律的条件下，使被控系统的状态变量跟踪参考模型的状态变量，且两种模型不需要具有相同的结构和参数。运用λ范数证明了算法的收敛性，仿真结果验证了算法的有效性。

关键词: 　非线性, 模型参考 , 　迭代学习控制 ,    , λ范数