The servo system control precision is difficult to improve because of the presence of nonlinear friction. Therefore, the servo system model with the dynamic LuGre friction is established. The adaptive fuzzy controller is designed in the case of unknown system parameters and load torque. The adaptive fuzzy logic system is used to approximate the nonlinear part including LuGre friction to achieve high-precision position tracking. It can be ensured theoretically by Lyapunov function that the closed-loop system is stable. Simulation results show that the controller can compensate the influence of nonlinear friction, and be robust for the load torque change.