Abstract
We argue that endowing machines with perception and action raises many problems for which solutions derived from first principles are unfeasable or too costly and that artificial neural networks offer a worthwhile and natural approach towards a solution. As a result, learning methods for creating mappings or dynamical systems that implement a desired functionality replace the implementation of algorithms by programming. We discuss two neural network approaches, the Local Linear Maps (LLMs) and the Parametrized Self-Organizing Maps (PSOMs), that are well suited for the rapid construction of mapping modules and illustrate some of their possibilies with examples from robot vision and manipulator control. We also address some more general issues, such as the need for mechanisms of attention control and for the flexible association of continuous degrees of freedom, together with an ability for handling varying constraints during the selection of actions.
Preview
Unable to display preview. Download preview PDF.
References
Brady M. (ed.): Robotics Science. MIT Press, Cambridge Massachusetts (1989)
Daugman, J.G.: Two-dimensional spectral analysis of cortical receptive field profiles, Vision Research 20 (1980) 847–856
Erwin E., Obermayer K., Schulten K.: Models of Orientation and Ocular Dominance Columns in the Visual Cortex: A Critical Comparison. Neural Computation 7 (1995) 425–468
Heidemann G., Ritter H.: A Neural 3-D Object Recognition Architecture Using Optimized Gabor Filters. Proc. 13th Int. Conf. Patt. Recog., Vol. IV, (1996) 70–74, IEEE Computer Society Press.
Heidemann G., Ritter H.: A neural recognition architecture for composed objects. DAGM Symposium Mustererkennung (B. Jähne, P. Geißler, H. Haußecker, F. Hering eds.), (1996) 475–482, Springer Verlag Berlin Heidelberg.
Heidemann, G., Kummert F., Ritter H., Sagerer G.: A Hybrid Object Recognition Architecture, ICANN 96, Springer Lecture Notes in Computer Science 1112, (1996) 305–310. Springer Berlin Heidelberg.
Heidemann, G., Nattkemper T., Menkhaus G., Ritter H.: Blicksteuerung durch präattentive Fokussierungspunkte. Proceedings in Artificial Intelligence (B. Mertsching ed.) (1996) 109–116 Infix Verlag, St. Augustin (in German).
van Hemmen, J.L., Kühn R.: Collective Phenomena in Neural Networks. In: Models of Neural Networks (E. Domany et al. eds.) 1–106, Springer Berlin Heidelberg 1991.
Hunt K.J., Sbarbaro D., Zbikowski R., Gawthrop P.J.: Neural Networks for Control Systems: A Survey. Automatica 28 (1992) 1083ff.
Klein C.A., Huang C-H.: Review of Pseudoinverse Control for Use with Kinematically Redundant Manipulators. IEEE Trans. Sys. Man and Cybern. SMC13, (1983) 245–250.
Kohonen, T.: The Self-Organizing Map, Proc. IEEE 78 (1990) 1464–1480
Kohonen, T.: Self-Organizing Maps, Springer Series in Information Sciences, Springer Berlin Heidelberg 1997.
Littmann E., Drees A., Ritter H.: Neural Recognition of Human Pointing Gestures in Real Images. Neural Processing Letters (1996) 61–71, Kluwer Academic Publishers.
Littmann E., Ritter H.: Adaptive Color Segmentation-A comparison of Neural and Statistical Methods. IEEE Trans. Neural Networks 8 (1997) 175–185
Maes P. (ed.): Designing Autonomous Agents (Special Issue of Robotics and Autonomous Systems), MIT Press Cambridge (1990)
Meyering A., Ritter H.: Learning 3D-Shape Perception with Local Linear Maps. Proc. Int. Joint Conf. on Neural Networks (1992) IV:432–436, Baltimore
Moratz R., Heidemann G., Posch S., Ritter H., Sagerer G.: Representing Procedural Knowledge for Semantic Networks using Neural Nets. Proc. 9th Scand. Conf. Image Anal. (1995) 819–828 Uppsala.
Niemann H., Sagerer G., Schröder S., Kummert F.: ERNEST: A Semantic Network for Pattern Understanding. IEEE Transactions on Pattern Analysis and Machine Intelligence 12 (1990) 883–905 Nöllcer, C., Ritter, H.
Nöllcer, C., Ritter, H.: Detektion von Fingerspitzen in Videobildern DAGM Mustererkennung (F. Wahl et al. ed.) (1997) (accepted) Springer Verlag Berlin Heidelberg.
Obermayer K., Ritter H., Schulten K.: A Model for the Development of the Spatial Structure of Retinotopic Maps and Orientation Columns. IEICE Trans. Fundamentals E75-A, (1992) 537–545.
Pearlmutter B.: Gradient calculation for dynamic recurrent neural networks: a survey. IEEE Trans. Neural Networks 6 (1995) 1212–1228
Press, W.H., Flannery, B.P., Teukolsky, S.A., Vetterling, W.T.: Numerical Recipes in C, Cambridge University Press (1990).
Ritter H., Schulten K.: Extending Kohonen's Self-Organizing Mapping Algorithm to Learn Ballistic Movements. In: Neural Computers (Eckmiller R., v.d.Malsburg Ch. eds.) (1987) 393–406 Springer Heidelberg
Ritter, H., Martinetz, T., Schulten, K.: Topology Conserving Maps for Learning Visuomotor-Coordination. Neural Networks 2 (1989) 159–168.
Ritter, H., Martinetz, T., Schulten, K.: Neural Computation and Self-organizing Maps, Addison-Wesley, Reading, MA. (1992)
Ritter, H. Parametrized Self-Organizing Maps. ICANN 93 Proceedings (S.Gielen and B.Kappen eds.), (1993) 268–273, Springer Berlin Heidelberg.
Ritter H.: Learning with the Self-Organizing Map. In: Artificial Neural Networks (T. Kohonen et a1. ed.) (1991) 379–384.
Ritter H.: Self-Organizing Feature Maps: Kohonen Maps. In: The Handbook of Brain Theory and Neural Networks (M.A. Arbib ed.) 846–851 MIT Press, Cambridge (1995)
Walter J., Ritter H.: Investment Learning with Hierarchical PSOM. In: Advances in Neural Information Processing Systems 8 (NIPS*95) (D. Touretzky, M. Mozer and M. Hasselmo eds.) (1996) 570–576, MIT Press
Walter J., Ritter H.: Local PSOMs and Chebychev PSOMs-Improving the Parametrized Self-Organizing Maps. ICANN95 Conf. Proceedings (1995) 95–102
Walter J.: Rapid Learning in Robotics. Cuvillier Verlag Göttingen 1996.
Walter, J., Ritter, H., Rapid Learning with Parametrized Self-organizing Maps. Neurocomputing 12, (1996) 131–153
Yeung D., Bekey G.: On Reducing Learning Time in Context Dependent Mappings, IEEE Transactions on Neural Networks 4 (1993) 31–42
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Ritter, H. (1997). Neural network approaches for perception and action. In: Sommer, G., Koenderink, J.J. (eds) Algebraic Frames for the Perception-Action Cycle. AFPAC 1997. Lecture Notes in Computer Science, vol 1315. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0017878
Download citation
DOI: https://doi.org/10.1007/BFb0017878
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63517-8
Online ISBN: 978-3-540-69589-9
eBook Packages: Springer Book Archive