In the context of the position and speed information of the rotor being difficult to directly obtain in the position sensorless control of permanent magnet synchronous motors, this paper establishes a PMSM position sensorless vector control system model and designs a speed and position estimation structure consisting of an extended Luenberger observer and NLESO-PLL. This method first uses the extended Luenberger observer to estimate the motor’s back electromotive force, and then estimates its components in the α and β coordinate systems from the stator voltage and current signals. Then, the estimated back electromotive force is input into NLESO-PLL, and the rotor position, angular velocity, and mechanical speed are estimated through nonlinear error feedback and the expansion state observation loop. Compared with the traditional PLL, NLESO-PLL can enhance dynamic tracking capability and anti-disturbance performance under conditions of sudden speed changes and load disturbances. Using MATLAB/Simulink, a PMSM position sensorless vector control simulation model is built, and speed step, load disturbance, and position estimation experiments are conducted under different operating conditions. The simulation results show that the proposed method can better track the actual speed and rotor position in the medium-to-high-speed operating range and maintain good estimation stability under speed changes and load disturbances, providing a feasible solution for position-sensorless control of PMSM at medium and high speeds.