ZHANG NEURAL NETWORK FOR LINEAR TIME-VARYING EQUATION SOLVING AND ITS ROBOTIC APPLICATION

摘要：

Different from gradient-based neural networks, a special kind of recurrent neural network has recently been proposed by Zhang et al for real-time matrix inversion.In this paper, we generalize such a design method to solving online a set of linear time-varying equations.In comparison with gradient-based neural networks, the resultant Zhang neural network for time-varying equation solving is designed based on a vector-valued error function, instead of a scalar-valued error function.It is depicted in an implicit dynamics, instead of an explicit dynamics.Furthermore, Zhang neural network globally exponentially converges to the exact solution of linear time-varying equations.Simulation results, including the application to robot kinematic control, substantiate the theoretical analysis and demonstrate the efficacy of Zhang neural network on linear time-varying equation solving, especially when using a power-sigmoid activation function.

关键词： Time-varying linear equations recurrent neural network error function implicit dynamics power-sigmoid activation function robot kinematic control

作者: YU-NONG ZHANG HAI-FENG PENG

作者单位: Department of Electronics and Communication Engineering,Guangzhou 510275, China School of Life Science Sun Yat-Sen University, Guangzhou 510275, China

会议类型: 国际会议

会议名称: 2007 International Conference on Machine Learning and Cybernetics(IEEE第六届机器学习与控制论国际会议)

会议地点: 香港

会议语种:英文

页码: 3543-3548

在线出版日期: 2007-08-19（万方平台首次上网日期，不代表论文的发表时间）