Adaptive pole-placement control with multi-rate type neural network for pneumatic servo system

We propose a design scheme which combines an adaptive pole-placement control with a neural network (NN) for the pneumatic servo system. In this design scheme, we adopt a multi-rate type NN in which an update-time of weights is selected as multiple of a sampling-time. The effectiveness of the proposed design scheme is confirmed by experiments using the existent pneumatic servo system.