Authors: Victor Vladareanu; Paul Schiopu; Mingcong Deng; Hongnian Yu
Addresses: Faculty of Electronics, II, 'Politehnica' University of Bucharest, 313 Sp. Independenţei, Bucharest, Romania ' Faculty of Electronics, II, 'Politehnica' University of Bucharest, 313 Sp. Independenţei, Bucharest, Romania ' Department of Electrical and Electronic Engineering, Tokyo University of Agriculture and Technology, Koganei, Tokyo 184-8588, Japan ' School of Design, Engineering and Computing, Bournemouth University, Poole, Dorset BH12 5BB, UK
Abstract: The paper presents the design and simulation of a three degrees-of-freedom mechatronic system for use in an autonomous robotic platform. The tracking reference for the angular joints is given using a trained neuro-fuzzy model instead of the usual inverse kinematic system. The joint actuators are driven using a selection of position controllers featuring PID control, simple and enhanced extenics control and a number of fuzzy controllers, some of which recently introduced. A comparison for their performance is then provided. The paper draws on some previously investigated concepts of great actuality to the field such as neuro-fuzzy networks, extenics and reduced-base fuzzy control.
Keywords: robotic manipulators; neuro-fuzzy training; extenics control; reduced-base fuzzy control; walking motion; autonomous robot; legged locomotion; walking robots; robot control; mechatronics design; neural networks; fuzzy logic; joint actuators; position control; PID control; simulation; robot motion; motion control.
International Journal of Advanced Mechatronic Systems, 2015 Vol.6 No.4, pp.137 - 146
Received: 26 Nov 2014
Accepted: 02 Mar 2015
Published online: 25 Oct 2015 *