Maglev trains require accurate, real-time, and reliable speed measurement and positioning systems. However, traditional single-sensor systems suffer from limitations such as insufficient accuracy, susceptibility to interference, and error accumulation. To address these challenges, we propose an enhanced integrated positioning algorithm that builds upon the existing Interactive Multiple Model Kalman Filter (IMM-KF) by incorporating a fuzzy logic algorithm. Firstly, by analyzing the technical principles and performance characteristics of BeiDou satellite navigation system and inertial measurement device, the combined positioning technology architecture of magnetic levitation train with multi-sensor information fusion is constructed. Secondly, to address the problem of fixed measurement noise variance and model probability transfer matrix in the IMM-KF algorithm, the measurement noise variance adjustment factor and probability matrix correction factor are introduced, and the measurement noise variance and model probability transfer matrix are adjusted and corrected by fuzzy inference. Simulation experiments show that the algorithm proposed in this paper improves the positioning accuracy compared with the Adaptive Interactive Multiple Model-Adaptive Kalman Filter (AIMM-ARKF) algorithm and IMM-KF algorithm, respectively. Before and after the change of the maneuvering state of the maglev train, this algorithm can switch to the matching model more accurately.