In this article we review some intelligent and robotic algorithm approaches and propose a novel neural-network (NN) based approach for nonlinear systems. The nonlinear systems can be represented by the nonlinear Tagaki-Sugeno (T-S) and NN models. The linear differential inclusion state-space representation is utilized to deal with the controlled systems. The stability conditions and controller design for this representation are derived based on the fuzzy parallel distributed compensation scheme which is employed to construct a global fuzzy logic controller by blending all local state feedback controllers. The time-delay states that exist in the nonlinear simulated example, including the chaotic disturbances, are given to show the feasibility of the proposed fuzzy controller design approach.