Neural networks for control, robotics and diagnostics

Abstract

This special issue of Neural Computing and Applications (NCA) presents 12 original articles, which are extended versions of selected papers from the Fourth International Symposium on Neural Networks (ISNN2007), June 3–7, 2007, Nanjing, China. This prestigious annual conference is organized by Prof. Derong Liu from the Department of Electrical and Computer Engineering, University of Illinois at Chicago and is technically co-sponsored by the IEEE Nanjing Section, IEEE Computational Intelligence Society, International Neural Networks Society, Asia Pacific Neural Networks Assembly and European Neural Networks Society. The contributions of this issue reflect the well-known fact that ISNN traditionally covers a broad variety of the thoroughness of techniques deployed for control, robotics and diagnostics of neural networks. Based on the recommendation of the special session organizers and the reviews of the conference papers, a number of authors were invited to submit an extended version of their conference paper for this special issue of NCA. All the invited articles were thoroughly reviewed once again by at least two independent experts and, finally, the 12 articles presented in this volume were accepted for publication. Papers were selected on the thoroughness of techniques deployed rather than the basis of fundamental ideas/ concepts. In this special issue 12 papers are included. 1. In the paper ‘‘Near Optimal Neural Control of Multiple-Input Nonlinear Systems’’ by Dingguo Chen, Jiaben Yang and Ronald R. Mohler, a new neural network-based optimal control is proposed to deal with a special class of uncertain nonlinear systems with multiple inputs. 2. In the paper ‘‘Neural Adaptive Control for a Class of Nonlinear Systems with Unknown Deadzone’’ by Zhonghua Wang, Yong Zhang and Hui Fang, a deadzone precompensator is developed using backstepping design techniques. Transient performance is guaranteed and stability is obtained. 3. In the paper ‘‘Neurodynamic Programming: A Case Study of the Traveling Salesman Problem’’ by Jia Ma, Tao Yang, Zeng-Guang Hou, Min Tan and Derong Liu, two methods, temporal difference learning and approximate Sarsa, are presented to solve the large-scale traveling salesman problem based on neurodynamic programming. 4. In the paper ‘‘RBFN-based Decentralized Adaptive Control of a Class of Large-Scale Non-affine Nonlinear Systems’’ by Tong Zhao, a radial basis function neural network (RBFN) adaptive control scheme is proposed to control a class of large-scale decentralized nonlinear systems with strong interconnections. 5. In the paper ‘‘Multiple Models Switching Control Based on Recurrent Neural Networks’’ by Jun-Yong Zhai, Shu-Min Fei and Xiao-Hui Mo, the transient performance of nonlinear discrete-time systems is improved by generalized minimum variance controllers and compensating controllers. 6. In the paper ‘‘A Method for Condition Monitoring and Fault Diagnosis in Electromechanical System’’ by Qianjin Guo, Haibin Yu, Jingtao Hu and Aidong C. Sun (&) College of Electrical Engineering, Hohai University, 210098 Nanjing, China e-mail: cysun@hhu.edu.cn