Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
380246 | Engineering Applications of Artificial Intelligence | 2016 | 15 Pages |
A realistic control paradigm should concurrently account for different sources of uncertainty such as those in modeling parameters, external disturbances and noise, as well as operational cost. Yet, this is a daunting task for which many current control approaches lack in one aspect or the other. In particular, the consumed control energy is an important aspect of controller design that is often ignored. In this paper, we propose a stable robust adaptive interval type-2 fuzzy H2/H∞ controller (RAIT2FH2H∞C) for a class of uncertain nonlinear systems that aims to address the above concerns through its hybrid robust/adaptive structure. In particular, the H2 energy and tracking cost function is minimized with respect to a H∞ disturbance attenuation constraint, while the adaptive interval fuzzy logic system (IT2FLS) handles the uncertainties in approximating the unknown nonlinear dynamics of the system. In principle, the interval fuzzy logic approach aims to manage portions of uncertainty that could not be precisiated, leading to improved error performance. Several simulation studies, with or without disturbance and noisy measurements, as well as actual experimental implementation on a 3-PSP (prismatic-spherical-prismatic) parallel robot confirm this assessment. More specifically, in comparison with a competing methodology as well as its type-1 counterpart, the proposed interval type-2 strategy expends better or comparative control effort while reaching considerably better tracking performance.