I. Introduction
In 1991, Chern presented the integral variable structure control (IVSC) to solve the steady-state error problem and enhance the robustness of the traditional variable structure control (VSC) [1]. Chern then applied this control scheme to a robot manipulator [2], brushless dc servo system [3], [4], [8], dc motor [5], and induction motor [6], [9], [11] to demonstrate the feasibility of IVSC.
Functional schematic of CMAC.
Related research has also proposed a method of designing IVSC [7]. Other applications include the engine system [10], the uninterruptible popwer supply (UPS) system [13], and the voltage regulator system [12]. The IVSC solved the steady state error problem of conventional VSC due to the dead zone or the boundary layer by introducing the integral state variable into the controller design [14], [15]. However, regardless of whether the traditional VSC or the IVSC scheme is employed, the controller design is a parametric scheme that depends strong on the system model or parameters. In the past decade, however, the nonparametric scheme fuzzy sliding mode control (FSMC) was proposed and outperforms the traditional VSC scheme that needs the system parameters [27] [28]– [30]. FSMC retains the favorable aspects of SMC but improves robustness by integrating the IF-THEN linguistic rule into the SMC controller, using the expert's experience of systematic tuning scheme. However, the need for human expertise or a complex tuning methodology remains an outstanding difficulty.