Almost sure convergence of ILC for networked linear systems with random link failures

The iterative learning control (ILC) problem for networked linear system with random link failure is addressed in this paper. The link failures arise both from the controller to the plant and from the plant to the controller. The random link failure is modeled by an independent Bernoulli random variable. Two P-type update laws are designed and critically analyzed. The almost sure convergence property is established according to this case for the first time. Illustrative simulations show the effectiveness of the proposed algorithms.