Title: Inverse kinematics of manipulator using weighted fuzzy clustering method for fuzzy training data

Authors: Vijayant Agarwal, B.C. Nakra, A.P. Mittal

Addresses: Industrial Automation and AI Lab, Netaji Subhas Institute of Technology, Sec.03, Dwarka, New Delhi – 078, India. ' Mechanical & Automobile Engineering Dept., Institute of Technology and Management, Sector 23 A, Gurgaon – 122017, Haryana, India. ' Instrumention and Control Engineering, Netaji Subhas Institute of Technology, Sec.03, Dwarka, New Delhi – 078, India

Abstract: The inverse kinematics of redundant manipulator is considered. A model-free regression approach based on weighted fuzzy clustering method is formulated. For the adopted technique, the observed or training data pair is fuzzy instead of crisp for known value of joint variables to enhance the practicability of inverse kinematics solutions, since the real-time data collected by the sensors is generally fuzzy or vague instead of crisp. Simulation results indicate that this method has higher identifying precision and better real-time ability. Therefore, a new way for solving the inverse kinematics of manipulator for fuzzy data is proposed.

Keywords: inverse kinematics; redundant manipulators; weighted fuzzy clustering; fuzzy data; training data; regression; redundant robots; simulation.

DOI: 10.1504/IJAISC.2009.027290

International Journal of Artificial Intelligence and Soft Computing, 2009 Vol.1 No.2/3/4, pp.176 - 187

Published online: 19 Jul 2009 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article