Run-to-run control with nonlinearity and delay uncertainty

In this paper, we study a simple single-input single output nonlinear system controlled by a Run-to-run algorithm. Besides the usually considered model uncertainty, the particularity of the system under consideration is that measurements available to the control algorithm suffer from large and varying measurement delays. The control algorithm is a nonlinear sampled model-based controller with successive model inversion and bias correction. The main contribution of this article is its proof of global convergence. In particular, the model error and the varying delays are treated using monotonicity of the system and a detailed analysis of the closed-loop behavior of the sampled dynamics, in an appropriate norm.