Robustness of discrete-time iterative learning control for networked control systems with data dropouts

This paper investigates the performance of iterative learning control (ILC) scheme that is adopted in networked control systems with data dropouts, where a linear discrete-time stochastic system can be rewritten as a super-vector formulation. In this paper, two types of compensation schemes are employed for data dropouts occur in both the output signals and control input signals during the signals transmission from the sensor to the ILC controller and from ILC controller to the actuator, respectively. Through statistical analysis approach, it is shown that ILC can perform well and achieve bounded convergence in the sense of norm. Numerical simulations demonstrate the validity and effectiveness.