Most available results on iterative learning control address trajectory tra
cking problem for systems without time delay. The effect of time delay on t
racking performance is not yet fully understood. This paper is concerned wi
th iterative learning control design for the systems with delayed states, w
here the system class under consideration is confined by a defined relative
degree. A learning control scheme is proposed to overcome the uncertaintie
s in delay times and /or in system parameters. Robustness of the learning c
ontrol is established in the presence of initial function errors. Furthermo
re, uniform convergence of system outputs to the desired trajectory is ensu
red if the initial function at each cycle is set to align with the desired
initial function. Validity of the results is demonstrated through numerical
simulations.