Date on Master's Thesis/Doctoral Dissertation


Document Type

Doctoral Dissertation

Degree Name

Ph. D.


Electrical and Computer Engineering

Committee Chair

Lilly, John H.

Author's Keywords

Sliding mode control; Pneumatic muscles; Robot manipulators


Manipulators (Mechanism)


This dissertation is concerned with investigating robust approaches for the control of pneumatic muscle systems. Pneumatic muscle is a novel type of actuator. Besides having a high ratio of power to weight and flexible control of movement, it also exhibits many analogical behaviors to natural skeletal muscle, which makes them the ideal candidate for applications of anthropomorphic robotic systems. In this dissertation, a new phenomenological model of pneumatic muscle developed in the Human Sensory Feedback Laboratory at Wright Patterson Air Force Base is investigated. The closed loop stability of a one-link planar arm actuated by two pneumatic muscles using linear state feedback is proved. Robotic systems actuated by pneumatic muscles are time-varying and nonlinear due to load variations and uncertainties of system parameters caused by the effects of heat. Sliding mode control has the advantage that it can provide robust control performance in the presence of model uncertainties. Therefore, it is mainly utilized and further complemented with other control methods in this dissertation to design the appropriate controller to perform the tasks commanded by system operation. First, a sliding mode controller is successfully proposed to track the elbow angle with bounded error in a one-Joint limb system with pneumatic muscles in bicep/tricep configuration. Secondly, fuzzy control, which aims to dynamically adjust the sliding surface, is used along with sliding mode control. The so-called fuzzy sliding mode control method is applied to control the motion of the end-effector in a two-Joint planar arm actuated by four groups of pneumatic muscles. Through computer simulation, the fuzzy sliding mode control shows very good tracking accuracy superior to nonfuzzy sliding mode control. Finally, a two-joint planar arm actuated by four groups of pneumatic muscles operated in an assumed industrial environment is presented. Based on the model, an integral sliding mode control scheme is proposed as an ultimate solution to the control of systems actuated by pneumatic muscles. As the theoretical proof and computer simulations show, the integral sliding mode controller, with strong robustness to model uncertainties and external perturbations, is superior for performing the commanded control assignment. Based on the investigation in this dissertation, integral sliding mode control proposed here is a very promising robust control approach to handle systems actuated by pneumatic muscles.