A fuzzy-smooth variable structure filtering strategy: For state and parameter estimation

Abstract

This article introduces a new filtering strategy based on combining elements of fuzzy logic and the smooth variable structure filter (SVSF). A revised formulation of the SVSF is presented in an effort to combine it with fuzzy logic, and is referred to as the RSVSF. Computer simulations are used to compare the new strategy, referred to as the Fuzzy-SVSF, with other popular Kalman-based estimation methods. Preliminary results indicate that combining fuzzy logic with the SVSF yields an improved estimation result and improved stability to system and modeling changes and uncertainties. © 2013 IEEE.