Fuzzy controller design using a new shuffled frog leaping algorithm

Abstract

The paper presents a new shuffled frog leaping algorithm (SFLA) for optimal design of fuzzy controllers. The SFLA is a meta-heuristic iterative search method inspired from the memetic evolution of a group of frogs when seeking for food. It consists of a frog leaping rule for local search and a memetic shuffling rule for global information exchange. In this paper, a new frog leaping rule with exponentially decaying uncertainties is proposed to enhance performances of the SFLA. With this new frog leaping rule, the SFLA has a wide local search space to prevent premature convergence at the first iterations, and a narrow local search space to accelerate convergence speed at the latter iterations. The new SFLA is then applied to design fuzzy controllers such that a specified performance criterion is minimized. The effectiveness of the proposed SFLA-based fuzzy controller is illustrated via an application to a ball and beam system.