Título: Interceptação Visual de Alvos 3D usando Redes Neurais de Aprendizado Não Supervisionado
Autores: Viana, Sidney Antonio A.; Nascimento Jr., Cairo L.; Waldmann, Jacques
Resumo: This paper concerns the application of two neural network architetures to solve the problem of visual interception of a stationary 3D target by a stereo (binocular) vision system. The neural networks are trained using competitive and unsupervised learning. The first neural architecture uses two Fuzzy-ART pattern-clustering neural networks which are combined with a linear “representation layer” to act as a fast open-loop neural controller and provides a rapid and coarse positioning of the visual system. The second neural architecture is a Kohonen network that acts as a closed-loop neural controller and provides a slower but more accurate positioning of the visual system. We show how to use these two neural architectures separately and combined to solve the visual interception problem. Simulation results of some interception tasks are presented and show that the better results are obtained when both neural architectures are combined.
Artigo em pdf: 4cbrn_047.pdf
Arquivo BibTex: 4cbrn_047.bib