Abstract:A decentralized neural network based position control method is presented for multi-fingered dexterous hand with nonlinear tendon-drive system. Multiple decentralized neural network is acquired via learning nonlinear relationship of tendon-driving system using the simple input output data that obtained by applying different input to the complex drive system. Then the inverse model of the tendon-drive system can be established using the decentralized neural networks according to the characteristics of the tendon-drive system. In the meantime, an online neural network is used to keep well running state for a long time. Experiment illustrates the proposed method can improve greatly the tracking accuracy.
[1] Vincent Hayward and Juan Manuel Cruz-Hernandez.Parameter Sensitivity Analysis for Design and Control of Tendon Transmissions.Preprints of the Fourth International Symposium on Experimental Robotics Stanford,california,June 30-July 2,1995:1-11 [2] Stephen C Jacobsen,Hiekeun Ko,Edwin K.Iversen and Clark C Davis.Control Strategies for Tendon-Driven Manipulators.IEEE Control Systems Magzine,February 1990:23-28 [3] Makoto Kaneko,Mitsuo Wada,Hitoshi Maekawa,kazuo Tanie.A New Consideration on Tendon-Tension Control System of Robot Hands.Proceedings of the 1991 IEEE International Conference on Robotics and Automation Saramento Califomia-April,1991:1028-1033 [4] J H Lee,L W TASI.On the structural of tendon-driven manipulators having pseudo-triangular matrix.International J of Robotics and Research,1991,10(3): 255-262 [5] 王从庆,燕力欣,原魁,于达太.一种模糊学习控制方法及在多指手位置伺服系统中的应用.机器人,1995,17(2): 362-368 [6] 王国庆.基于神经网络的多指灵巧手控制.机器人,1996,5:311-315 [7] 李剑锋,张玉茹,张启先.多指灵巧手机构的速度操作灵巧性分析.机器人,1999,(2) [8] 张智星,孙春在,水谷英二.神经-模糊和软计算.西安交通大学出版社.2000