High-Speed Positioning of Pneumatic Robot Using Learning Posicast Control.

Abstract

In this study our objective is high-speed positioning of a pneumatic robot. To realize a high-speed drive, the control algorithm should be as simple as possible to reduce the calculation time of control input. A posicast control scheme may be considered an effective control strategy which satisfies such a requirement. However, the original scheme is not sufficient for a pneumatic drive system which is usually of an order higher than the second, and has an operating delay in the electro pneumatic valve, etc. We improve the original posicast control scheme so that it can be applied to positioning of the pneumatic robot. In this control scheme, the regulation of both the height and the switching time of reference input is important. Thus, we provide a learning function which can regulate these parameters with respect to the configuration of the last response during repeated operation. We describe the proposed control algorithm and discuss its availability through simulations and experiments.