具有初态学习的闭环PD型迭代学习控制

Closed-loop PD-type Iterative Learning Control with Initial State Learning

针对一类输入时滞非线性系统提出了一种新的学习控制算法,即在任意初始状态条件下系统的输入和初态同时进行学习的闭环PD型迭代学习控制,其中输入利用给定超前法。给出了该算法谱半径形式的收敛条件,并利用算子理论证明了系统在任意初始状态条件下经过迭代后,其输出能够完全跟踪期望轨迹。该算法解决了闭环PD型迭代学习控制的初始状态问题,且放宽了收敛条件。仿真结果表明了该算法的有效性。

For a class of nonlinear system with input time-dealy, a novel learning control algorithm is proposed, where the closed-loop D-type iterative learning control of an arbitrary initial state is conducted by the input and the initial state of the sys- tem simultaneously, where the input uses the given advance method. The convergence condition of the spectral radius form is given, and the operator theory is applied to prove that the output of the system with an arbitrary initial state can track the ex- pected trajectory completely after iteration. The problem of the initial state in the closed loop PD-type iterative learning control is solved, and the convegent condition is relaxed. The simulation results testify that the protposed algoruthm is effective.