sammut@qut.edu.au (David Sammut) (03/22/91)
Sorry to those who mailed me regarding the references for walking machines and the like but I lost my mail. So I posting them to the net below. Some of them are a little vague as my library does not have some of the journals and conference proceedings that are referenced. Thanks to all those that replied ! So here goes ... De Garis, Hugo, 1989 "WALKER, A Genetically Programmed, Time Dependent, Neural Net Which Teaches a Pair of Sticks to Walk", (Technical Report, Center for AI, George Mason Univ, Virginia) S.Y. Kitamura, Y. Kurematsu, and Y. Nakai , "Application of the Neural Network for the trajectory planning of a Biped locomotive Robot", Proc. of the First INNS Annual Meeting, Boston, Mass, 1988. J.J. Helferty, J.B. Collins, and M. Kam, "A Neural Network Learning Strategy for the Control of a One-Legged Hopping Machine", Proc. of the IEEE Intern. Conf. on Robotics and Automation, Scottsdale, Az, pp 1604 - 1609, May 1989. C.G. Atkeson, and D.J. Reinkensmeyer, "Using Associative Content- Addressable Memories to Control Robots", Proc. of the IEEE Intern. Conf. on Robotics and Automation, Scottsdale, Az, pp 1859 - 1864, May 1989. "Adaptive Control of a Legged Robot Using an Artificial Neural Network" Helferty, John J; J. Collins; M. Kam (Temple University) IEEE International Conference on Systems Engineering (Cat. No. 89CH2767-2) 1989 pp 165-8 "A Neural Network Learning Strategy for the Control of a One-Legged Hopping Machine" Helferty, John J; J. Collins; M. Kam (Temple University) 1989 IEEE International Conference on Robotics and Automation > The next reference might or might not be by Hugo De Garis (my library > doesn't have this either) but I have found some articles around that have been > written by him on the subject of Genetic Algorithms for Walking using NNs. GENETIC PROGRAMMING, Building Artificial Nervous Systems Using Genetically Programmed Neural Network Modules. Machine Learning, Proceedings of the Seventh International Conference on Machine Learning. Edited by Bruce W. Porter and Ray J. Mooney. >One of the replies was ... There was a talk on training a two-legged stickman to walk at the International Joint Conference on Neural Networks, January 1990, Washington, DC. I think it was done through genetic algorithms. I also think it was therefore in the "Biology track" of the conference. >I have been unable to check this out fully but I know it was done by >Genetic Algorithms (probably by Hugo De Garis again). Brooks, Rodney A., "A Robot that walks; Emergent Behaviors from a Carefully Evolved Network", Neural Computation Volume 1, No. 2 Summer (This also references a number of other articles by the same author) I am also aware of a book called "Making Them Move" (I don't know the author or publisher) which goes through related aspects of simulating movement such as for graphics and kinematic representations (which are handy for training neural networks). While it is not entirely devoted to walking machines and making them walk, it might be of interest. Thanks again to all those who contributed to this list and my apologies to those who requested this list whose mail I lost. If you would like an updated version in a few weeks please mail me again. If I can verify the exact references and sources for some of the above entries I will repost this as well as sending it to those parties that are interested. David Sammut +-----------------------------------------------------------------------+ | David Sammut | | | School of Computer Science | | | Faculty of Information Technology | email: sammut@qut.edu.au | | Queensland University of Technology | | | Brisbane, Queensland, Australia | | +-----------------------------------------------------------------------+