Dissertation
Generation Human Body Motion by the Centralized Networks
Abstract
The main goal of the thesis is to describe and study reduced models for efficient simulations of human body motions (HBM). To this end, we propose new coupled oscillator models, which are networks of dynamically coupled elements. These networks consist of a few of centers and many satellites. The centers evolve in time as periodical oscillators with different frequencies. The satellite states are defined via center states by a radial basis function (RBF) networks. To simulate different motions we adjust the parameter of the RBF networks and frequencies of centers. Moreover, our network includes a switching module that allows to turn from a motion to another one, say, from jumping to walking. It is shown that this model is capable to simulate complicated motions consisting of many different elements dynamical primitives (walking, running, jumping etc.).
Citation
@phdthesis{doi:10.17170/kobra-202311028942,
author={Morozov, Ivan},
title={Generation Human Body Motion by the Centralized Networks},
school={Kassel, Universität Kassel, Fachbereich Mathematik und Naturwissenschaften, Institut für Mathematik},
year={2023}
}
0500 Oax 0501 Text $btxt$2rdacontent 0502 Computermedien $bc$2rdacarrier 1100 2023$n2023 1500 1/eng 2050 ##0##http://hdl.handle.net/123456789/15206 3000 Morozov, Ivan 4000 Generation Human Body Motion by the Centralized Networks / Morozov, Ivan 4030 4060 Online-Ressource 4085 ##0##=u http://nbn-resolving.de/http://hdl.handle.net/123456789/15206=x R 4204 \$dDissertation 4170 5550 {{Körper}} 5550 {{Bewegung}} 5550 {{Neuronales Netz}} 5550 {{Radiale Basisfunktion}} 5550 {{Hopfield-Netz}} 5550 {{Simulation}} 7136 ##0##http://hdl.handle.net/123456789/15206
2023-11-22T16:10:27Z 2023-11-22T16:10:27Z 2023 doi:10.17170/kobra-202311028942 http://hdl.handle.net/123456789/15206 eng Urheberrechtlich geschützt https://rightsstatements.org/page/InC/1.0/ 510 Generation Human Body Motion by the Centralized Networks Dissertation The main goal of the thesis is to describe and study reduced models for efficient simulations of human body motions (HBM). To this end, we propose new coupled oscillator models, which are networks of dynamically coupled elements. These networks consist of a few of centers and many satellites. The centers evolve in time as periodical oscillators with different frequencies. The satellite states are defined via center states by a radial basis function (RBF) networks. To simulate different motions we adjust the parameter of the RBF networks and frequencies of centers. Moreover, our network includes a switching module that allows to turn from a motion to another one, say, from jumping to walking. It is shown that this model is capable to simulate complicated motions consisting of many different elements dynamical primitives (walking, running, jumping etc.). open access Morozov, Ivan 2023-09-06 iv, 82 Seiten Kassel, Universität Kassel, Fachbereich Mathematik und Naturwissenschaften, Institut für Mathematik Seiler, Werner M. (Prof. Dr.) Valkulenko, Sergey (Prof. Dr.) Regensburger, Georg (Prof. Dr.) Lindner, Felix (Prof. Dr.) Körper Bewegung Neuronales Netz Radiale Basisfunktion Hopfield-Netz Simulation publishedVersion false true
The following license files are associated with this item:
:Urheberrechtlich geschützt