The existing sensorless rotor position estimation technique for a dual-three phase permanent magnet synchronous motor (DTP-PMSM) that relies on the fundamental PWM excitation is able to achieve rotor position estimation at low and zero speeds without injecting high-frequency signals. However, it needs to take multiple current samples during individual active voltage vectors generated by a voltage source inverter. Although it is effective, the requirement on the multiple current sampling hinders its practical application. This paper therefore proposes a new algorithm to overcome the above shortcoming. The algorithm directly utilizes the saliency effect of dual three-phase PMSMs to extract the rotor position information in the average current derivatives of two adjacent PWM periods from the stationary coordinates by the synchronous sampling technique. The new algorithm allows a synchronous current sampling for both the rotor position estimation and the current control, which greatly eases the practical implementation. Finally, the proposed algorithm is effectively verified in a DTP-PMSM simulation model at low and zero speeds.