Vahid Pourfarzaneh - Department of Computer, Buinzahra Branch, Islamic Azad University, Buinzahra, Iran
Mahdi Mollamotalebi - Department of Computer and Information Technology Engineering, Qazvin Branch, Islamic Azad University, Qazvin, Iran
Ahmad Fakharian - Faculty of Electrical, Biomedical and Mechatronics Engineering, Qazvin Branch, Islamic Azad University, Qazvin, Iran

This paper proposes a formation control technique for leader-follower robots using ANFIS (Adaptive Neuro-Fuzzy InferenceSystem) that reduces the overhead compared to the neural network. It considered four mobile robots; one of them acts as aleader, and other ones are working as the followers. The user predefines the initial position of these robots. Using the proposedtechnique based on an adaptive neuro-fuzzy system, the robots are formed as the proper shape after limited iterations. Eachfollower robot adapts to the leader with regard to the leader’s speed and angle. Then, the follower robots adjust their distanceto the leader. It is assumed that there is no feedback from the followers into the leader; they locate according to the position ofthe leader and by considering the communication topology. The proposed technique is simulated in MATLAB and the resultvalues indicate higher performance of robots’ function compared to previous methods, CA and NFC. Finally, the proposedmethod is implemented in a real robotic platform and the experimental results show effectiveness of the proposed method inreal environment.

Formation control, Leader-follower, ANFIS, Iteration