Repetitive tracking control of DC motors using a fuzzy iterative learning controller

In this paper, we first study the theoretical design of a fuzzy iterative learning controller for sampled-time linear time invariant systems. Then we investigate the digital circuit implementation of the proposed controller with application to repetitive position tracking control of DC servo motors. The stability and convergence of the learning system are analyzed under uncertainties of initial state errors, input disturbance and output measurement error. We show that the learning error will converge to a residual set whose level of magnitude will depend on the size of the uncertainties. The learning error will asymptotically converge to zero if all the uncertainties disappear. To improve the learning performance, a concept of fuzzy learning gain is introduced which is designed based on the tracking error in the current and past iterations. In addition to the theoretical analysis, the fuzzy iterative learning controller is realized by a digital circuit to prove its feasibility. VHDL is used as the circuit design tool. The designed circuit is then downloaded into an FPGA chip. The chip is then applied to a repetitive position tracking control of DC motors to verify the learning effect. According to the experimental results, we prove that the theoretical design and its digital circuit are correct and feasible.