Introduction <p>Piezoelectric actuators (PEAs) are widely used in micro-platform vibration suppression for micron-level precision. However, inherent nonlinear hysteresis and external disturbances degrade control accuracy, restricting high-precision applications. Developing effective control strategies to mitigate these effects is a key research focus.</p> Purpose <p>To address the low accuracy and poor anti-disturbance ability of PEAs caused by nonlinearity and disturbances, this study proposes a high-performance compound control method to improve the trajectory tracking accuracy, response speed and robustness of piezoelectric micro-platforms, especially for unmanned vehicle-mounted scenarios.</p> Methods <p>A feedforward-feedback compound control strategy is designed. The feedforward module uses a PI-BiRNN hybrid model to compensate for nonlinear hysteresis via its inverse solution. The feedback module adopts a fractional order sliding mode control (FOSMC) based on power approximation law to suppress residual errors and external disturbances.</p> Results <p>Experiments on the piezoelectric micro-platform show that the proposed compound control outperforms traditional single feedforward/feedback control and existing compound schemes in trajectory tracking accuracy and disturbance suppression, with improved response speed and robustness. When mounted on an unmanned vehicle, it effectively compensates micro-vibrations, enhancing image clarity and system performance.</p> Conclusions <p>The compound control method effectively mitigates the adverse effects of PEA nonlinearity and disturbances. It provides a reliable technical solution for improving piezoelectric micro-platform performance, facilitating PEA applications in high-precision scenarios like unmanned vehicle-mounted micro-vibration suppression.</p>