Improved Control of a Pneumatic Actuator Pulsed with PWM

It is known that achieving accurate control for pneumatic systems is a challenging problem. This paper studies two effective algorithms for improving the performance of a pneumatic actuator. Without building the plant model, the proportional velocity acceleration (PVA) control with friction compensation is implemented to reduce the steady-state error. To overcome the time-delay problem unresolved by PVA, a CARIMA model-referenced adaptive generalized predictive control (AGPC) is applied for the pneumatic actuator, whose parameters are updated by on-line recursive least squares (RLS) method with forgetting factor. The experimental results of AGPC for the position control of the pneumatic actuator are impressive for both the steady-state and dynamic tracking