Sliding mode controller(SMC) is a simple but powerful nonlinear controller. However, to avoid high frequency chattering on the control input and guarantees stability, the sliding mode gain and the thickness of the boundary layer need to be chosen carefully. This paper proposes a new sliding mode controller with self-tuning the thickness of the boundary layer and the sliding mode gain. It uses a fuzzy rule base for tuning both the thickness of the boundary layer and the sliding mode gain to guarantee stability, to reject a discontinuous control input, and to improve the tracking performance. Design and control examples using the proposed method for a rotary inverted pendulum are performed to show the properties of the developed self-tuning algorithms. controller with self-tuning the thickness of the boundary layer and the sliding mode gain. It uses a fuzzy rule base for tuning both the thickness of the boundary layer and the sliding mode gain to guarantee stability, to reject a discontinuous control input, and to improve the tracking performance. With this scheme, the stability and the robustness of the proposed fuzzy logic control algorithm are proved and ensured by the sliding mode control law. Because the proposed algorithm tunes both the thickness of a boundary layer and the sliding mode gain for sliding mode controller, the algorithm has advantages that the tracking performance can be improved, good for rejection of control chattering phenomenon, and fairly large parameter variation and disturbances can be handled. A computer simulation example using the proposed method for a rotary inverted pendulum is performed to show the properties of the developed self-tuning algorithms.