分析了声纳图像拼接失真的因素，提出了一种基于仿生机械云台的声纳图像拼接方法. 根据眼球前庭动眼反射机理建立了一个具有主动补偿功能的仿生机械云台控制系统，对扫测艇姿态变化进行动态补偿减小了声纳图像拼接的误差. 使用姿态传感器与GPS以及图像信息对声纳图像进行对准，实现声纳图像的拼接. 不同环境下的仿真实验证明，该系统不仅具有动态补偿能力而且还具备鲁棒性. 最后，经水下声纳图像拼接的实验证明了该方法的有效性，较未使用仿生云台的拼接方法有更好的拼接效果.

This paper proposes a mosaicing method of sonar images based on a bionic mechanical pantilt system. From the principle of vestibule-ocular reflex, an active compensation control system of bionic mechanical pan-tilt is established. The system can compensate the sonar image instability due to attitude variation of the sweeping survey boat or submarines during operation. To obtain sonar image mosaics, the sonar images are matched with information of the attitude sensors and GPS. Simulation experiments are performed in different environments. The results show that the model is robust and has the ability of dynamic compensation. The results show effectiveness of the system.