A study was conducted on the low efficiency, poor accuracy, and reliance on experience of traditional manual docking methods for fixed wing aircraft cabin segments, with the aim of developing an automated cabin docking system based on multimodal data fusion. By using multimodal data real-time acquisition technologies such as vision, distance measurement, and force sensors, combined with servo control and AI image processing algorithms, a three-stage automated docking process including axis alignment, radial docking, and control feedback was constructed, achieving precise adjustment and real-time monitoring of cabin pose. The hardware system design integrates high-precision sensors, a movable adjustable platform, and a comprehensive control board, and develops main control software to achieve multimodal data fusion and closed-loop control. Through automated docking tests, the system has successfully completed high-precision docking of different shaped cabin segments, significantly improving docking efficiency, reducing manual intervention by more than 90%, lowering collision risks, and meeting the engineering application requirements under complex working conditions. The results indicate that the system effectively improves docking consistency and reliability, providing an efficient solution for modular assembly of aircraft and suitable for docking tasks of different types and sizes of cabin segments. In addition, in response to the limitations of the existing system, directions have been proposed to further optimize the multi degree of freedom adjustment function.