Robust Disturbance Observer-Based Adaptive Fuzzy Controller for Pneumatic Muscle Actuators

Variants of Fuzzy logic controllers (FLC) have been widely used to control the systems characterized by uncertain and ambiguous parameters. Control objectives for such systems become more challenging when they are subjected to uncertain environments. Human-robot interaction is such phenomenon wherein robot control difficulties are further augmented due to human intervention. State of the art of research in FLC has been limited in establishing a trade-off between accuracy and interpretability, since achieving both these performance measures simultaneously is difficult. In the present research, an adaptive FLC has been designed in order to achieve better accuracy and higher interpretability. Supported by another instance of FLC as disturbance observer, the proposed controller has adaptive mechanism specifically designed to alter its parameters. The adaptive FLC has been implemented to control actuation of a pneumatic muscle actuator (PMA). Experimental results show excellent trajectory tracking performance of the PMA in the presence of varying environment.