A recurrent network for data driven human movement generation

Author

Ciungradi, Bogdan ; Costa, Mario ; Pasero, Eros ; Macchiarulo, Luca

Author_Institution

Lab. Neuronica, Politecnico di Torino, Italy

Volume

1

fYear

1998

fDate

4-8 May 1998

Firstpage

216

Abstract

The problem of simulation and modelling of natural human movement is a basic topic in ergonomics applications. Generalization capabilities of artificial neural networks can be profitably pursued in this field. The use of real collected data for training may help in simulating human movement for CAD applications. Furthermore, it also permits us to embed features of the real movements which could be hardly taken into account with a priori modelling. We show an application of this point of view, consisting in recurrent networks trained to perform arm movements based on actual human data

Keywords

CAD; biomechanics; ergonomics; generalisation (artificial intelligence); learning (artificial intelligence); recurrent neural nets; data driven human movement generation; ergonomics; generalization capabilities; natural human movement; recurrent network; training; Artificial neural networks; Computational intelligence; Computational modeling; Ergonomics; Humans; Laboratories; Neural networks; Performance evaluation; Training data; Very large scale integration;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks Proceedings, 1998. IEEE World Congress on Computational Intelligence. The 1998 IEEE International Joint Conference on

Conference_Location

Anchorage, AK

ISSN

1098-7576

Print_ISBN

0-7803-4859-1

Type

conf

DOI

10.1109/IJCNN.1998.682265

Filename

682265