The point of this article is the sensor fault estimation and fault-tolerant controller design for IT2 fuzzy systems via sliding mode approach. In order to estimate accurately the system states and sensor faults, a novel proportional and derivative sliding mode observer is introduced, which provide more design freedom and eliminate the effects of sensor faults. By splitting the operating domain and estimating the membership functions, a set of relaxation stability conditions subject to the information of membership functions and system states are obtained. Then, a sliding mode controller in form of IT2 fuzzy model is designed to stabilize the closed-loop systems. An example of bolt-tightening tool model is considered to demonstrate the validity of the proposed results.