Abstract: In this paper, information filtering fusion theory is introducted into the field of low-speed maglev train speed and position detection. Beside the cross induction coil, radar speed sensors and balise are also adopted. The relative positioning is realized by the radar and cross induction coil, and the absolute positioning is realized by the balise. The information fusion comes true by this way. The accuracy and reliability of speed and position detection can be significantly improved. The final conclusion is verified by MATLAB simulation.