Trajectory tracking control of a pneumatic muscle system using fuzzy logic

Pneumatic Muscle (PM) system was first developed by McKibben in the 1950's as soft actuators in artificial limbs and became commercially available in the 1980's. They have since been used as actuators in high-tech robotic applications and in physical therapy for functional recovery since they are extremely safe in human presence compared to electric and hydraulic actuators. A high power to weight and power to volume ratios make the PM a very light yet powerful actuator. However, PM is highly nonlinear in nature due to its construction and mechanical properties and hence, they are difficult to control using a linear controller. Fuzzy logic is a good nonlinear modeling approach, since it uses fuzzy rules to handle nonlinearities. A fuzzy logic based feedforward controller (inverse dynamics) and feedback linearizing control schemes are proposed in this paper. Controllers are designed using data obtained from the PM system, and the design does not require a mathematical model. The PM parameters can change over time and varying operating conditions. Hence, an adaptive fuzzy algorithm is used to tune the fuzzy models to capture the parameter changes. The proposed control schemes are tested for its trajectory tracking capabilities and are found to yield excellent results.