A static walking robot to implement a neural network system

• Autorzy: Daya B. , Chauvet P.
• Języki publikacji: EN

Abstrakty

EN

In this paper, we consider the control of a simple static walking robot. We develop a prototype which will be able to test much architecture of artificial neural networks. The mechanism is driven by two servomotors. Its control system comprises an electronic circuit based on a programmable IC. The locomotion of the prototype has six states: forward, turn right, turn left, back right, pivot back and stay. We describe here the mechanism, the characteristics of the drives and the control strategy. In the first experiment which consists in avoiding obstacles, the designed algorithm was successfully implemented in the microcontroller, thus, making the robot intelligent.

Słowa kluczowe

EN

neural network system; microcontroller; intelligent robot; collision avoidance; static walking robot

• Wydawca: Oficyna Wydawnicza Politechniki Wrocławskiej
• Czasopismo: Systems Science
• Rocznik: 2005
• Tom: Vol. 31, no 1
• Strony: 27--35
• Opis fizyczny: Bibliogr. 13 poz., rys., wykr.

Twórcy

autor

Daya B.

• Institut Universitaire de Technologie (Lebanese University), B.P. 813, Sai'da, Lebanon

autor

Chauvet P.

• Insitut de Mathematiques Appliquees (IMA), Angers, France

Bibliografia

• [1] Chauvet P., Daya B., Chauvet G. A., Conditions of stability for the simulation of a neuromimetic circuit. Application to a bipedal robot, 12th European Simulation Multiconference, ESM'98, Manchester, United Kingdom, June 1998.
• [2] Daya B., Chauvet G. A., Control of dynamic biped locomotion using a neural network, WCNN96, INNS Press, San Diego 1996.
• [3] Daya B., A Multilayer Perceptrons Model for the Stability of a Bipedal Robot, Neural Processing Letters, Vol.9, 1999,221-227.
• [4] Franklinj A., Selfridge O. G., Sotne new directions for adaptive control theory in robotics [in:] Neural Networks for Control, (eds.) W. T. Miller III, R. S. Sutton and P. J. Werbos, Cambridge (MA): MIT Press, 1991, 349-360.
• [5] Ng G. W., Application of Neural Networks to Adaptive Control of Nonlinear Systems, John Wiley & Sons, 1997.
• [6] Hirasawa K., Kim S.-H., Hu J., Murata J., Han M., Jin C., Improvement of generalization ability for identifying dynamical systems by using universal learning networks. Neural Networks, Vol. 14, 2001, 1389-1404.
• [7] Houk J. C., Singh S. P., Fisher C., Barto A. G., An adaptive sensorimotor network inspired by the anatomy and physiology of the cerebellum [in:] Neural Networks for Control, (eds.) W. T. Miller 111, R. S. Sutton and P. J. Werbos, Cambridge (MA): MIT Press, 1991, 301-348.
• [8] Kagami S. et al.. Design and development of a Legged Robot Research Platform JROB-I, IEEE International Conference on Robotics and Automation, Leuven, Belgium, May, 1998, 146-151.
• [9] Pham D. T., Parhi Daya R., Navigation of multiple mobile robots using a neural network and a Petri Net model, Robotica, Vol. 21, 2003, 79-93.
• [10] Still S. et al.. Four-legged walking gait control using a neuromorphic chip interfaced to support vector learning algorithm, Advances in Neural Information Processing Systems. MIT Press, Cambridge, MA, 2000.
• [11] Thach W. T., Goodkin H. P., Keating J. G., The cerebellum and the adaptive coordination of movement, Annu. Rev. Neurosci., Vol. 15, 1992, 403-442.
• [12] Vukobratovic M., BOROVAC B., SuRLA D., Stokic D., Biped Locomotion, Springer-Verlag, Berlin 1990.
• [13] Yu-Yung C., Fu L., Yu-Chien C., Multi-agent Based Dynamic Scheduling for a Flexible Assembly System, IEEE Int'l Conf. on Robotics & Automation, Leuven, Belgium, May, 1998, 2122-2128.