Robust self-organising fuzzy sliding mode-based path-following control for autonomous underwater vehicles

Abstract

A robust self-organising fuzzy sliding mode control law steers autonomous underwater vehicles (AUVs) to track a predefined planar path at a constant speed without temporal specifications. An intelligent methodology has been adopted for path-following control to handle varying parametric uncertainties in vehicle dynamics and also conquers stringent preliminary condition constraints in several path-following control strategies illustrated in the literature. Robust controller design builds on a fusion of sliding mode control theory and fuzzy logic technique with an adaptation mechanism to tune boundary layer width and hitting gain. This novel strategy proposes two distinct tuning procedures: the first method commonly uses absolute error and their derivative as fuzzy input variables in a two-dimensional fuzzy logic rule structure. Herein, skew symmetry property is utilised in rule base structure to derive a single input fuzzy variable based on the signed distance technique, drastically reducing two-dimensional fuzzy logic rules. Since the second method provides substantial reductions in rule inferences through the use of the fuzzy rule's mirror image and the Lyapunov approach for tuning purposes, the resulting guidance control law yields fast convergence of the path-following error trajectory towards zero along with the elimination of chattering problem. Simulation results illustrate the effectiveness and robustness of the proposed control law to achieve favourable tracking performance with a high accuracy.