Abstract
An approach to visuomotor control using an artificial neural network (ANN) is presented. The architecture of the controller is founded on the Cerebellar Model Arithmetic Computer (CMAC) and consists of two principal elements: the Image CMAC (ICMAC), which provides a visual representation of a target object, and a Differential Image CMAC (DICMAC), which supplies rate information on the object. The approach differs from the conventional use of artificial vision in control in that the visual images are not merely reduced to position and orientation of the object but rather are employed integrally. Thus the controller would be able to generalize across a set of images and their corresponding objects. Learning is accomplished by means of a reinforcement scheme. Computer simulation results are presented for an inverted pendulum system.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
Albus J.S. (1981), Brain, behavior and robotics, Peterborough, N.H.: BYTE Books.
Baito A.G., Sutton R.S., Anderson C.W. (1983), Neuronlike elements that can solve difficult learning control problems. IEEE Transactions on Systems, Man, and Cybernetics, 13:835–846.
Graham D.P.W., D'Eleuterio G.M.T. (1991), MOVE-A Neural-Network Paradigm for Robotic Control, Canadian Aeronautics and Space Journal, V37, N1, pp 17–26.
Hashimoto H., Kubota T., Kudoli M., Harashima F. (1992), Self-Organizing Visual Servo System Based on Neural Networks, IEEE Control Systems, April, pp 31–36.
Kuperstein M. (1990), INFANT Neural Controller for Adaptive Sensory-Motor Coordination, Neural Networks, V4, pp 131–145.
Martinets T., Ritter H.J., Schulten K.J. (March 1990), Three-Dimensional Neural Net for Learning Visuomotor Coordination of a Robot Arm, IEEE Transactions on Neural Networks, V1, N1, pp 131–136.
McGuire P.F., D'Eleuterio G.M.T. (1992), Active Control of Interference in CMAC/MOVE Neural Networks for Robotic Applications, Seventh CASI Conference on Astronautics, Ottawa, ON, 4–6.
Miller W.T., Glanz F.H., Kraft L.G. (1987a), Application of a General Learning Algorithm to the Control of Robotic Manipulators, The International Journal of Robotics Research, V6, N2, pp 84–98.
Miller W.T. (1987b), Sensor based Control of Robotic Manipulators using a general Learning Algorithm., IEEE Journal of Robotics and Automation, V3, N2, pp 157–165.
Pomerleau (1993), Neural network perception or Mobile Robot Guidance, Carnegie Mellon University: Kluwer Academic Publishers.
Sutton R.S. (1988), Learning to predict by methods of temporal difference, Machine Learning, V3, pp 9–44.
Watkins CJ.C.H. (1989), Learning from Delayed Rewards, PhD thesis, Kings College.
Watkins C.J.C.H. and Dayan P. (1992), Q-learning, Machine Learning Journal, 8(3/4). May 1992 Special Issue on Reinforcement Learning.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1995 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
McGuire, P.F., D'Eleuterio, G.M.T. (1995). Visuomotor control using an artificial neural network. In: Mira, J., Sandoval, F. (eds) From Natural to Artificial Neural Computation. IWANN 1995. Lecture Notes in Computer Science, vol 930. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-59497-3_287
Download citation
DOI: https://doi.org/10.1007/3-540-59497-3_287
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-59497-0
Online ISBN: 978-3-540-49288-7
eBook Packages: Springer Book Archive