This paper proposes a new method to minimize torque ripple of a switched reluctance motor (SRM), which is based on the control of the sum of the square of the phase currents by using only two current sensor and analog multipliers. In addition, the sliding mode control (SMC) technique has been applied to SRM speed control loop that compensates the low frequency oscillations on the torque output. The applied sliding mode controller adjusts the value of the reference current to keep the speed of the motor constant. The proposed sliding mode continuous controller performs better than PI and Fuzzy controllers and is also more robust to the external disturbances. The results show that the continuous sliding mode controller is effective compared to PI or Fuzzy controllers in reducing the torque ripple of the motor, compensating for the nonlinear torque characteristics and making the drive insensitive to parameter variations as well. Since, mutual inductances have an important disturbance effect on the torque ripple, mathematical model of the SRM has been derived in presence of them. It has been used a completely analog system to implement the proposed algorithm. Also, a mechanical sensor has been used to sense the rotor position. Simulation and experimental results are given to validate the proposed method for the self and mutual inductances and current measurements.