Comparison between three types of iterative learning control for nonlinear systems: On-line, off-line and on-line taking into account the previous error

This paper deals with iterative learning control “ILC” to solve the trajectory tracking problem for nonlinear continuous-time systems having a non linear output. Thus, a comparison between three types of this control scheme has been made: on-line ILC, offline ILC and on-line ILC taking into account the previous error. The λ -norm is adopted as the topological measure in our proof of the asymptotic stability of these control schemes, over the whole finite time-interval, when the iteration number tends to infinity. Simulation results on a nonlinear system showed us that on-line ILC taking into account the previous error is better in terms of speed of convergence that off-line and on-line ILC.