Hysteresis exists widely in intelligent materials, such as piezoelectric and giant magnetostrictive ones, and it significantly affects the precision of vibration control when a controlled object moves at a range of micrometers or even smaller. Many measures must be implemented to eliminate the influence of hysteresis. In this work, the hysteresis characteristic of a proposed piezoelectric actuator (PEA) is tested and modeled based on the adaptive neuro fuzzy inference system (ANFIS). A linearization control method with feedforward hysteresis compensation and proportional–integral–derivative (PID) feedback is established and simulated. A linear quadratic Gaussian with loop transfer recovery (LQG/LTR) regulator is then designed as a vibration controller. Verification experiments are conducted to evaluate the effectiveness of the control method in vibration isolation. Experiment results demonstrate that the proposed vibration control system with a feedforward feedback linearization controller and an LQG/LTR regulator can significantly improve the performance of a vibration isolation system in the frequency range of 5–200 Hz with low energy consumption.