Immune evolution algorithm for iterative learning controller

In this paper, immune evolution algorithm (IEA) by imitating the defending process of an immune system and the mutating ideas of biology evolutionary is investigated to optimize the input of iterative learning controller. In the IEA, a self-adaptive mutation operator is constructed to decide the mutation step size of every antibody by its environment and an affinity calculation process is also embedded to maintain the diversity. The method takes the objective function that is defined as the square error between reference signal and output signal in all sampling points and constraints as antigen. Through the genetic evolution, an antibody that most fits the antigen becomes the solution. The experimental results confirm that the proposed method has higher tracking accuracy and fast convergence speed. And compared with conventional iterative learning control methods, it is easy to solve the optimal input for nonlinear plant models.