Rule-base generation via symbiotic evolution for a Mamdani-type fuzzy control system

Abstract

In this paper, an intelligent-based active suspension system using a Mamdani-type fuzzy logic controller is developed. In order to formulate the rule-base, a new algorithm based on symbiotic evolution is proposed. Because almost all fuzzy rule-base generation algorithms usually produce a structure with redundant and overlapped membership functions, an algorithm that merges such similar membership functions is also integrated within this approach. It is shown that this algorithm leads to a more transparent and more interpretable rule-base with a minimum number of membership functions and a reduced number of rules. As a test-bed an active suspension system was used that was based on a quarter-car model with parameters relating to a Ford Fiesta MK2. The proposed method was compared with a PID (proportional-integral-derivative)-based active suspension system whose parameters were optimized using genetic algorithms (GAs). Simulation results show that the symbiotic evolution-based fuzzy controller achieved better performances in all carried-out investigations.