Yin, S.; Xu, W., and Yu, C.L., 2020. Direct torque control of permanent magnet synchronous motor in ship propulsion system based on memory attenuation Kalman filter. In: Al-Tarawneh, O. and Megahed, A. (eds.), Recent Developments of Port, Marine, and Ocean Engineering. Journal of Coastal Research, Special Issue No. 110, pp. 80–86. Coconut Creek (Florida), ISSN 0749-0208.Aiming at the problem that the control system filter in ship electric propulsion system is easy to diverge, this study adopted the Memory Attenuation Kalman Filter (MAKF) to design a direct torque controller for permanent magnet synchronous motor (PMSM). By detecting the voltage and current of the stator, this study achieved the direct observation of phase current, rotate speed and rotor position angle signal, and the indirect observation of torque and flux linkage via MAKF; this method can accurately and stably estimate information such as the rotate speed of the motor and the position of the rotor, and it has overcome the problem of Kalman Filter divergence. The simulation and test results showed that the state variables of the motor observed by this method were accurate, the maximum error of the transient state rotate speed estimation was only 8%; in terms of astable estimation error, the rotor position signal lag was only 0.04T; when the load torque changed suddenly, the motor ran steadily and responded quickly, showing good robustness and reliability.