Abstract: In the permanent magnet synchronous linear motor (PMSLM) system, when the incremental sensors and sensorless technology are adopted in the closed-loop control, the existence of detent force and friction will cause a large deviation in initial position estimation, affecting the starting performance of the system. In view of this problem, a pre-positioning estimation algorithm considered the detent force and friction of PMSLM system was proposed. Based on the analysis of the detent force and friction characteristics of the system, the dynamic balance relationship of the motor during the process of pre-positioning was analyzed. The estimated achievable constraints of the equilibrium position of pre-positioning were then derived, and an iterative estimation algorithm of the equilibrium position of pre-positioning based on the dichotomy method was proposed in the stationary coordinate system. The simulation and experiment of the pre-positioning estimation algorithm were carried out, which verified the feasibility and effectiveness of the algorithm via a flat-type PMSLM in cooperation with the drive control platform of "ARM+FPGA".