A new fuzzy filter with application in motion control systems

We propose a fuzzy logic-based filter. Different from existing fuzzy filters, our model applies the self-organizing map algorithm together with the least squares method to get the optimal membership functions, as well as the consequent inference parameters. This approach results in a delayless fuzzy filter that has more efficient filtering capability than conventional fuzzy filters. A DC servo motor system is employed as a testbed for this fuzzy filter. Simulation shows that our fuzzy filter can efficiently filter out the harmful velocity noise produced by the low-cost tachometer in the feedback loop.