Abstract :A new scheme of sensorless control of an induction motor is proposed in which soft2computing methodologies are used in the paper. The T2S fuzzy method of online stator resistance estimation of an induction motor is described ,where the re2 sistance value is derived from stator2windings temperature estimation as a function of stator current and frequency. The speed and online rotor resistance of transient state are estimated by the basic MRAC ,and elman neural network is used to formthe current2 speed model . The online rotor resistance of steady state is set to change with the same rate of stator resistance value change. The scheme has been designed and iterated by simulation study and the performance in low speed is improved. Keywords :induction motor speed sensorless control soft2computing