In this paper, in order to synthesize a control law we propose a new approach that enables identification of the intermediate equilibrium points of a nonlinear system, knowing the first and the last ones. These points are those around which the nonlinear system is linearized and therefore yields local models (sub-models) that contribute to forming the multimodel describing the nonlinear system. This approach is based on the transition from a given point (source) to the next by varying a scheduling parameter (SP) defining the source point sub-model. The variation of this parameter is limited by the maximum value of the stability margin determined by the loop shaping design procedure approach (LSDP) applied to such a sub-model. Hence, the new equilibrium point is defined by the new obtained value of the SP for which the gap metric between this sub-model and the one corresponding to the new value of SP is larger than the given stability margin. The different robust controllers synthesized for the different equilibrium points will be used to synthesize the robust control of the nonlinear system, by applying the gain-scheduling technique. The proposed transition approach as well as the robust control algorithm were validated on the continuous stirred tank reactor (CSTR) system.