Precise control of a pneumatic muscle actuator is a challenging problem, due to its nonlinear and time-varying characteristics. Specifically, it can be modeled as parallel connection of a nonlinear friction force, a nonlinear spring, and a nonlinear contractile element. It is difficult to identify the coefficients associated with these elements within a certain precision as they change along the course of continuous use. In this paper, we introduce a novel adaptive servomechanism controller and rigorously prove that the trajectory tracking can be achieved with a desirable error bound. Moreover, an experiment platform is set up to demonstrate the effectiveness of the proposed controller in real applications. In particular, the proposed adaptive controller significantly improves the tracking accuracy and achieves better performance than non-adaptive and PID controllers and other existing controllers in literature.

No abstract