<p>This paper presents a reliable strategy to control and estimate the speed of synchronous variable reluctance motor (SynVRM) in fuel cell (FC) electric vehicles. The fuel cell ensures the electrical supply of the SynVRM via a converter-inverter system. A fractional-order PI (FOPI) controller is used to minimize the overshoot of the responses in the different closed loops, in one hand. In the other hand, to ensure the operation continuity in the event of speed sensor failure, a switched speed tracking algorithm (SSTA) is proposed to detect the error signal from the speed sensor. Thus, a model reference adaptive system (MRAS) observer is used to replace the speed sensor and to estimate the real speed, thus switching from the normal control mode to a sensorless control mode. In this context, the main goals and the innovation of this paper include: 1) to ensure a robust control even in the failure case of speed sensor, 2) to guarantee optimal operation of all parts of the electrical fuel cell vehicle. The simulations show the performances and the efficiency of the proposed fractional-order PI controller for both control and speed estimation. Indeed, the proposed FOPI controller is more efficient than the classic PI controller, whether in dynamic or static behavior. Moreover, the proposed SSTA algorithm allows continuous operation of the control system during speed sensor failure at constant and variable speed.</p>