This paper presents a new technique for linearized systems with uncertainties using sliding mode control with fuzzy tuning. The control signal consists of three components: the equivalent control for assigning the sliding eigenvalues of the nominal closed-loop system, the robust control for driving the representative point to the sliding hyperplane in the reaching phase, and the fuzzy tuned control for accelerating the reaching time and reducing the influence of unstructured uncertainty. Chattering is also taken into account. The validity of the proposed approach is verified by the control of a two-mass servo system. Simulation results show that the controlled system is highly robust against parameter and load variations, and backlash nonlinearity.