World Applied Sciences Journal 22 (12): 1782-1788, 2013 ISSN 1818-4952 © IDOSI Publications, 2013 DOI: 10.5829/idosi.wasj.2013.22.12.2606 Corresponding Author: Mickael Aghajarian, Department of Electrical & Computer Engineering, Semnan University, Semnan, Iran. Tel: +98 (913) 3051982, E-mail: m_aghajarian@semnan.ac.ir. 1782 Design of Fuzzy Controller for PUMA 560 Robot Arm Using Improved Bacterial Foraging Optimization Algorithm Mickael Aghajarian and Kourosh Kiani Department of Electrical and Computer Engineering, Semnan University, Semnan, Iran Abstract: Trial and error method can be used to find a suitable design of a fuzzy controller. Generally, the design of fuzzy controller involves determination of the fuzzy rules, Membership Functions (MFs) and scaling factors. An optimization algorithm facilitates the design process and finds an optimal design to achieve a desired performance. This paper presents an Improved Bacterial Foraging Optimization Algorithm (IBFOA) to design a fuzzy controller for tracking control of a PUMA 560 robot arm driven by permanent magnet DC motors. We use efficiently the IBFOA to form the rule base and MFs. To show the improvement of proposed algorithm, the IBFOA is compared with Bacterial Foraging Optimization Algorithm (BFOA) and Particle Swarm Optimization (PSO) algorithm. Performance of the controller in the joint space and in the Cartesian space is evaluated. Simulation results show superiority of the IBFOA to the BFOA and PSO algorithm. Key words: Evolutionary Algorithms Fuzzy Logic PUMA 560 Robot Arm Tracking Control INTRODUCTION not be solved using conventional problem solving A wide variety of control strategies were proposed to algorithms like Bacterial Foraging Optimization Algorithm control robot manipulators. PID controls are certainly the (BFOA) [6], Particle Swarm Optimization (PSO) [7] and Ant most widely adopted control strategy in industry because Colony Optimization (ACO) [8] have been dominating the of its simple structure and robust performance in a wide realm of optimization algorithms and proved their range of operating conditions. Although PID control effectiveness. offers the simplest and yet most efficient solution to many Since a foraging organism or animal takes a real world control problems [1], optimally tuning gain necessary action to maximize the energy obtained per unit is quite difficult [2]. Alternatively, fuzzy control as a time spent for foraging, in the face of constraints model-free approach is simply designed to control presented by its own physiology, such as sensing and complicated systems [3]. To form fuzzy rules, an exact cognitive capabilities, natural foraging strategy can be knowledge of model is not required. Fuzzy controller is an applied to real-world optimization problems. Based on intelligent controller using linguistic fuzzy rules to include such evolutionary idea, Passino proposed BFOA as an information from experts. Consequently, fuzzy control of optimization algorithm [6]. BFOA is a new evolutionary robot manipulators has attracted a great deal of computation technique, which also includes powerful researches to overcome uncertainty, nonlinearity and optimization techniques like PSO [7] and ACO [8]. coupling by providing a model-free control [4]. To design To improve BFOA search performance, several a Fuzzy Logic Controller (FLC), a major task is to researchers have extended the basic BFOA to deal with determine fuzzy rules, Membership Functions (MFs) multi-modal and high dimensional functions [9-11]. and scaling factors. Therefore, the controller is tuned Researchers are trying to hybridize BFOA with other until a desired performance is achieved. An optimization evolutionary algorithms [12-13] in order to reduce the algorithm facilitates the design process and finds an convergence time and enhance the accuracy. Over certain optimal design to achieve a desired performance. In the real-world optimization problems, BFOA has been last decade, approaches based on evolutionary reported to outperform many powerful optimization algorithms have received increased attention from the algorithms like genetic algorithm [14] and PSO researchers for solving optimization problems that could algorithms [15]. techniques [5]. Recently natural swarm inspired