Title: Genetic algorithms with variable search space function for fine gain tuning of model-based robotic servo controller

Authors: Akimasa Otsuka; Akihiko Ikeda; Fusaomi Nagata

Addresses: Department of Mechanical Engineering, Faculty of Engineering, Tokyo University of Science, Yamaguchi, Daigaku-Dori 1-1-1, Sanyo-Onoda, Yamaguchi, 756-0884, Japan ' Department of Mechanical Engineering, Faculty of Engineering, Tokyo University of Science, Yamaguchi, Daigaku-Dori 1-1-1, Sanyo-Onoda, Yamaguchi, 756-0884, Japan ' Department of Mechanical Engineering, Faculty of Engineering, Tokyo University of Science, Yamaguchi, Daigaku-Dori 1-1-1, Sanyo-Onoda, Yamaguchi, 756-0884, Japan

Abstract: In this paper, genetic algorithms with a variable search space function are proposed for fine gain tuning of a resolved acceleration controller which is one of model-based robotic servo controllers. To realise a stable controller, position and velocity feedback gains should be tuned suitably. Proposed genetic algorithms temporally vary the search space if a certain condition is satisfied. The variable search space function is activated in genetic algorithms if the optimal solution is not updated for fixed generations. When the function is active, an update within the fixed generations or non-updated situation for the fixed generations makes the function terminated. The proposed method is evaluated through a trajectory following control problem, in which it is tried to search better feedback gains more rapidly. Simulations are conducted by using the dynamic model of PUMA560 manipulator. The results demonstrate the effectiveness and the promise of the proposed method.

Keywords: fine gain tuning; genetic algorithms; variable search space; resolved acceleration control; PUMA560 manipulator; model-based control; robot control; servo control; trajectory following; trajectory control; simulation; dynamic modelling.

DOI: 10.1504/IJMMS.2013.052782

International Journal of Mechatronics and Manufacturing Systems, 2013 Vol.6 No.1, pp.23 - 37

Received: 31 Mar 2012
Accepted: 08 Oct 2012

Published online: 12 Jul 2014 *

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