Kohonen Network Application to the Problem of a Manipulator Working Area Modeling
Abstract
The issue of using artificial neural network technologies in problems of computer modeling of robotics objects is considered using the example of constructing and clustering the working area of jointly working manipulators. Kohonen network structure is determined and training without a teacher is performed. A comparative analysis is carried out with the results obtained by solving the direct kinematics problem.
About the Authors
S. K. Shulgin
Lugansk State University named after Vladimir Dal
Russian Federation
Russian Federation
Shulgin S.K., Candidate of Technical Sciences, Assistant Professor.
D. O. Sinepolsky
Lugansk State University named after Vladimir Dal
Russian Federation
Russian Federation
Sinepolsky D.O., Senior Lecturer.
References
1. Шахинпур М. Курс робототехники. – М.: Мир, 1990. – 527 c.
2. Фу К., Гонсалес Р., Ли К.; Перевод с англ. А. А. Сорокина и др. под ред. В. Г. Градецкого. – М.: Мир, 1989. – 620 с.
3. Kohonen, T. Self-Organizing Maps, Berlin — New York: Springer-Verlag, 2001.
4. Уоссермен, Ф. Нейрокомпьютерная техника: Теория и практика = Neural Computing. Theory and Practice. — М.: Мир, 1992. — 240 с.
Review
For citations:
Shulgin S.K., Sinepolsky D.O. Kohonen Network Application to the Problem of a Manipulator Working Area Modeling. Social-economic and technical systems: research, design and optimization. 2024;(1):92-98. (In Russ.)
Views:
6
Email this article 