Authors:
Veljko Potkonjak
1
;
Nenad Bascarevic
1
;
Predrag Milosavljevic
1
;
Kosta Jovanovic
1
and
Owen Holland
2
Affiliations:
1
University of Belgrade, Serbia
;
2
University of Sussex, United Kingdom
Keyword(s):
Knowledge Base Control, Fuzzy Control, Machine Learning, Robot Arms, Bio-inspired Robot.
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Computational Intelligence
;
Fuzzy Control
;
Fuzzy Systems
;
Fuzzy Systems Design, Modeling and Control
;
Soft Computing
Abstract:
This paper aims to present a novel experience-based solution for a black-box control problem, applied to an anthropomimetic robot. The control method is tested on a point to point control problem of a multi-jointed robot arm. The model characteristics – dynamics, kinematics, and control parameters – are considered as unspecified, and therefore we deal with a machine learning approach that follows the cybernetic concept of black-box. The only available data of the system are those obtained from measuring inputs and outputs. The control algorithm involves two levels: feedforward and feedback. The main focus is, however, on feedback level where the algorithm for experience-based estimation of kinematic coefficients is combined with fuzzy logic control in order to relate the control inputs with the robot arm motion in the global frame.