Statistical proof of the adequacy of the neural network model of the executive kinematic chain of the manipulator transportable degrees of mobility
Abstract
The application of statistical methods for proving adequacy in the application to the neural network model of the manipulator is considered. The model of the manipulator kinematics is constructed using the matrix Denavit-Hartenberg method. The training sample is formed by solving the direct kinematics problem. The structure is selected and the neural network is trained to solve the inverse kinematics problem. Using the method for estimating the distribution of residuals, the adequacy of the resulting neural network model to the original kinematics model based on transcendental equations is proven.
About the Authors
S. K. Shulgin
Lugansk State University named after Vladimir Dahl
Russian Federation
Russian Federation
Candidate of technical Sciences, assistant professor
D. O. Sinepolsky
Lugansk State University named after Vladimir Dahl
Russian Federation
Russian Federation
Senior Lecturer
V. V. Makogon
Lugansk State University named after Vladimir Dahl
Russian Federation
Russian Federation
Assistant
References
1. Шахинпур М. Курс робототехники. – М.: Мир, 1990. – 527 c.
2. Фу, Кинсан. Робототехника : пер. с англ. / К. Фу, Р. Гонсалес, К. Ли; пер. А. А. Сорокин, А. В. Градецкий, М. Ю. Рачков ; ред. В. Г. Градецкий. - Москва : Мир, 1989. - 621 с.
3. R. Koker : “A genetic algorithm approach to a neural-network-based inverse kinematics solution of robotic manipulators based on error minimization,” Information Sciences, Vol. 222, pp. 528–543, 2013.
4. A.V. Duka : «Neural network based inverse kinematics solution for trajectory tracking of a robotic arm» Procedia Technology, Vol. 12, pp. 20–27, 2014.
5. Y. Maeda, T. Fujiwara and H. Ito : «Robot control using high dimensional neural networks» Procs. of SICE Annual Conference 2014, pp. 738–743, 2014.
6. Дрейпер Н., Смит Г. Прикладной регрессионный анализ. – М.: Финансы и статистика, 1987. – 350 с.
7. Кендалл М., Стьюарт А. Статистические выводы и связи – М.: Наука, 1973.– 900 с.
Review
For citations:
Shulgin S.K., Sinepolsky D.O., Makogon V.V. Statistical proof of the adequacy of the neural network model of the executive kinematic chain of the manipulator transportable degrees of mobility. Social-economic and technical systems: research, design and optimization. 2024;(3):110-119. (In Russ.)
Views:
4
Email this article 