Abstract—This paper proposed a fuzzy controller for the autonomous navigation problem of robotic systems in a dynamic and uncertain environment. In particular, we are interested in determining the robot motion to reach the target while ensuring their own safety and that of different agents that surround it. To achieve these goals, we have adopted a fuzzy controller for navigation and avoidance obstacle, taking into account the changing nature of the environment. The approach has been tested and validated on a Thymio II robots set. As application field, we have chosen a parking problem. Index Terms—Mobile robot, navigation autonomous, fuzzy logic, Thymio II robot. I. INTRODUCTION Obstacle avoidance is an essential component to achieve successful navigation [1]. Several research works have been reported in this area. Several trajectory tracking and path following algorithms have been proposed to steer the mobile robot along a path to a desired goal in order to prevent collisions with obstacles. Many researches turned their attention to the obstacle avoidance problem developing interesting real-time behavior for mobile robots in unknown environment. The most well know are, the potential field method, vector field histogram and the method of deformable virtual zone. The first one was introduced by [2] imagines the virtual forces acting on the robot. This method assumes that the robot is driven by virtual forces that attract it towards the goal, or reject it away from the obstacles. The actual path is determined by the resultant of these virtual forces, the second method is introduced in [3] which corresponds to local occupancy grid, constructed from the sensors of the robot; this method was improved in [4], landmark learning [5], edge detection, graph-based methods [6], Limit-cycles method [7] and many others. However, relatively few of them are suitable for real-time and embedded applications on very low-cost systems. Among them, the methods that use artificial intelligence approaches as:-Neural Network [8]: this approach is applied to determine the optimal neural networks structure for real- time obstacle avoidance task. In [9], the paper proposes a Manuscript received December 7, 2013; revised March 24, 2014. F. Boufera and Fatima Debbat are with the Mathematics and Computer Science Department, University of Mascara, Mascara, Algeria (e-mail: fboufera@yahoo.fr, debbat_fati@yahoo.fr). F. Mondada is with EPFL, Lausanne, Robotic Systems Laboratory (e- mail: francesco.mondada@epfl.ch). M. F. Khelfi is with Laboratory of Researches in Industrial Computing & Networks, Faculty of Exact and Applied Sciences, University of Oran, Algeria (e-mail: mf_khelfi@yahoo.fr). neural network that uses Qlearning reinforcement technique for solving the problem of obstacle avoidance. -Multi Agent System: In [10], a new local collision avoidance algorithm between multiple robots for real-time navigation is presented. This algorithm is based on multi agent system and quadratic optimization method for a collision free navigation and to compute the motion of each robot. - Hybrid Neural Network Genetic Algorithm: [11] used a hybrid neural net- work, genetic algorithm and local search method for solving the problem of finding the optimal collision free path in complex environments for mobile robot. -Particle Swarm Optimization: In [12], the author proposed two extensions of Particle Swarm Optimization (PSO) and Darwinian Particle Swarm Optimization (DPSO) to solve the problem of obstacle avoidance for multi robot. All these methods require to be implemented on robots with a sufficiently complex functional structure. The robot must possess capabilities of perception, decision, evaluation of the action and a relatively high computational power and/or memory. This paper mainly deals the navigation control and obstacle avoidance. Our objective is to develop a simple and reactive obstacle avoidance tool that can be implemented on an extremely compact such as the Thymio II [13]. In this context, we propose an intelligent and fast fuzzy controller system for navigating in real-time. The fuzzy logic is certainly one of the most adopted approaches in industry. It addresses such applications perfectly as it resembles human decision making with an ability to generate precise solutions from certain or approximate information. It fills an important gap in engineering design methods left vacant by purely mathematical approaches (e.g. linear control design), and purely logic-based approaches (e.g. expert systems) in system design. The advantage of using fuzzy logic for navigation is that it allows for the easy combination of various behaviors outputs through a command fusion process. The navigation system in this case consists of two behaviors an obstacle avoidance behavior and a goal seeking behavior [14], [15]. A set of experimentations is realized to demonstrate the feasibility of this approach for navigation, avoiding static and dynamic obstacles (other robots) and aggregation (in our case, parking in the edges of environment) for Thymio II mobile robots. Besides this introduction, the structure of the paper is as follows: Section II gives the specifications of the robot platform and its functioning under ASEBA tool. Section III presents the description of fuzzy logic controller. Section IV shows different experimentations for testing the proposed approach. Conclusions and perspective are given in Section V. Fatma Boufera, Fatima Debbat, Francesco Mondada, and M. Fayçal Khelfi Fuzzy Control System for Autonomous Navigation and Parking of Thymio II Mobile Robots 321 International Journal of Computer and Electrical Engineering, Vol. 6, No. 4, August 2014 DOI: 10.7763/IJCEE.2014.V6.846