Abstract: According to the basic principle of iterative learning control, a new approach to iterative learning control algorithms is proposed, which is based on Bang-Bang control theory. This method is different from the methods commonly used in that the later ones depend on the iterative errors or differential coefficient of errors. Its control law has just relation to the sign of iterative errors, which not only reduces the calculation but also improves the anti-disturbance performance. The expressions and control structure of the new algorithms are given. Then the convergence is analyzed. Simulations illustrate the effectiveness and convergence of the new algorithms.

Key words: iterative, learning control, Bang-Bang control