抄録
Sliding mode control is a well-known technique to guarantee robustness in the presence of uncertainties of modeling, parameter variations, and external disturbances. The sliding mode control law is based on the knowledge of controlled system and the norm (or maximum value, etc.) of uncertainties. However, the modeling work is difficult, and the cost of it is expensive. So, this paper proposes the self-tuning sliding mode controller that calculates the control input (manipulated variable) only by using the desired value and the state variable without requiring the system parameter, the input parameter and the size of the disturbance. Various experiments, which are the temperature control of aluminum and wood-ceramics, the level control of the water tank, and the position control of the shape memory alloy in the program control (time-scheduled control), show the validity and utility of the proposed controller.
