A DSP based sampled-data iterative learning control system for brushless DC motors

A sampled-data iterative learning controller using current error learning is presented in this paper. A detailed convergence and robustness analysis of the learning system at each sampling instant is studied. It is shown that learning error converges to a residual set whose level of magnitude depends on the bounds of uncertainties and the learning gain. In general, the learning speed is faster and learning error is reduced if the learning gain is larger. This learning algorithm is then applied to the position tracking control of brushless DC motors. A digital signal processor (DSP) based control board, which combines the power of the TI TMS320F243 DSP and the flexibility of Xilinx FPGAs, is used to develop the learning system. The nice learning performance shown in the experiment results proves the validity and practicability of the proposed learning controller.