1. INTRODUCTION Although fuzzy logic controller (FLC) was proposed almost three decades ago, and were successfully applied in many applications, but the accuracy of this controller depends on number of membership functions for input variables. That is, large number of membership functions usually guarantee small steady state error. On the other hand, large number of membership functions results in enormous number of fuzzy rules, which is referred to as the curse of dimensionality. This makes the design and application of the fuzzy controllers very cumbersome. In this paper a combined control law will be presented which has two modes: fuzzy- sliding mode and state-feedback mode. This combined control law takes advantage of both the fuzzy sliding and the state feedback. That is, High speed response due to the fuzzy-sliding part and small steady state error because of the state- feedback part. This controller works as follows: when the states are far from the desired states (or from the sliding surface) the fuzzy-sliding law is used, and when the states are near them, the control law uses the state feedback law. The reminder of this paper is organized as follows. Section 2 discusses the characteristics of the nonlinear magnetic ball suspension system (MBSS). In section 3 the classical sliding-mode controller will be designed for MBSS. In section 4 the fuzzy sliding- mode controller, with two different number of membership functions, will be given and the results will be compared with the classical sliding-mode controller. The combined controller (fuzzy-sliding and state feedback) is presented in section 5. Simulation results show the superiority of the proposed method as compared to the fuzzy sliding and the classical sliding-mode controllers. FUZZY-SLIDING STATE-FEEDBACK CONTROL OF NONLINEAR BALL SUSPENSION SYSTEM Saeed Mohammad-Hosseini, Ali-Reza Alfie Mohammad Farrokhi, and Mohammad-Reza Jahed-Motlagh Department of Electrical Engineering Iran University of Science and Technology Abstract: This paper presents a fuzzy-sliding control, based on state feedback methods, to control nonlinear magnetic ball suspension system. Simulations results show that the control speed of fuzzy-sliding-mode controllers is more than the classical sliding mode controllers, but the steady state error of later is less that the earlier. This is due to the limited number of membership functions defined for the input variables of the system. Here, a combination of fuzzy-sliding-mode and state-feedback control is presented. Simulation results show that this controller contains the advantages of both the fuzzy- sliding-mode method (fast response) and the state-feedback scheme (low steady-state error). Copyright  2002 IFAC Keywords: fuzzy control, magnetic suspension, state-feedback control, sliding-mode control Copyright © 2002 IFAC 15th Triennial World Congress, Barcelona, Spain www.Matlabi.ir