Hybrid Force-Position Dynamic Control of the Robots Using Fuzzy Applications Luige Vladareanu 1, a , Victor Vladareanu 2,b and Paul Schiopu 2,c 1 Institute of Solid Mechanics of the Romanian Academy, C-tin Mille 15, Bucharest 1, ROMANIA 2 “Politehnica” University of Bucharest, Faculty of Electronics II, 313 Splaiul Independenţei, 060042 Bucharest, ROMANIA a luigiv@arexim.ro, b vladareanuv@gmail.com, c schiopu.paul@yahoo.com Keywords: robot dynamic control, fuzzy control loops, hybrid force-position control, open architecture systems. Abstract. The paper presents methods of improving the hybrid force-position dynamic robot control using fuzzy logic for the error control. The implementation of the open architecture control system for robots using fuzzy application allows for the control of the hybrid position and force in Cartesian coordinates through real time processing of the Jacobean matrix obtained out of forward kinematics using the Denevit-Hartenberg method and calculating the Jacobean inverted matrix for control in closed loop. The effectiveness of various fuzzy control structures in controlling the force-position of the robot or mechatronics actuators is presented. Introduction The implementation of the open architecture control system for robots with compliant wrist allows for the control of the hibrid position and force in cartezian coordonates through real time processing of the Jacobine matrix obtained out of the forward kinematics using the Denevit- Hartenberg method and calculating the Jacobine inverted matrix for control in closed loop. Using the joint rate control reprezentation, having J() as the position Jacobean matrix and X as the generalized position error vector, this process is analyzed in a simultaneous two-way fashion: the first to determine the X F matrix corresponding to the force controled component and the second to determine the X P matrix corresponding to the position controled component. The  F joint error of the force component and the  P position component error are insered to a fuzzy controller. Fuzzy variables for input and output of the system and the membership function reflecting the deflection in Cartezian coodonates are studied. The history of fuzzy logic starts with the paper published in 1965 by L.A. Zadeh, entitled ‘Fuzzy sets’, in which the author introduces his new approach to set theory. Essentially, a fuzzy set is an extension of a classical bivalent (crisp) set with ‘a membership function which assigns to each object a grade of membership between 0 and 1’ (Zadeh, 1965) [1]. Subsequent detailed investigations made by Mamdani [2] and Takagi and Sugeno [3] have led to Fuzzy Logic becoming an increasingly appealing alternative to classical control for an array of systems (Thomas, Armstrong-Helouvry 1995) [4]. Jantzen (2007) [5] pioneers the development of algorithms for tuning the fuzzy rule base, beginning from a linear controller and the adjusting for non – linear behaviour. Further improvement in this area, which is arguably the most consequential for a fuzzy controller, should at least lead to improved industry acceptance and solve a key problem with fuzzy controllers today. In a similar way that the existence of Ziegler – Nichols contributes to the popularity of PID controllers by guaranteeing an acceptable working solution, a universally accepted general tuning algorithm for the fuzzy rule base exists would, in all probability, contribute to fuzzy controllers becoming the standard in non – linear control architectures. Applied Mechanics and Materials Vol. 245 (2013) pp 15-23 © (2013) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AMM.245.15 All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of TTP, www.ttp.net. (ID: 78.96.111.114-26/10/12,21:46:17)