In this paper, a new approach of linguistic fuzzy modeling proposed by Ben-Ghalia et al. is applied to induction motor. Thus, a classical model is first given together with the most used asynchronous motor control strategy, which is the pole assignment. After more, in order to minimize the dependence of the system to parameter variations and external perturbations, and reduce the number of variable states, a new model of the original system is proposed. This model is too simple and is suitable to cognitive approach such as fuzzy modeling. Then, fuzzy modeling of an uncertain system using the approach of Ben-Ghalia et al. to reduce the effects of imperfections of the classical modeling are introduced. In addition, their applications in the fuzzy sliding mode control of the asynchronous motor are presented. Finally, Simulation results reveal some very interesting features and show that the proposed fuzzy model has great potential for use as an alternative to the approximation of the conventional model of induction motor taking into account the effect of system model uncertainties and gives a good performances in sliding mode control.