Improving Neural Networks for Mechanism Kinematic Chain Isomorphism Identification

Detection of isomorphism among kinematic chains is essential in mechanical design, but difficult and computationally expensive. It has been shown that both traditional methods and previously presented neural networks still have a lot to be desired in aspects such as simplifying procedure of identification and adapting automatic computation. Therefore, a new algorithm based on a competitive Hopfield network is developed for automatic computation in the kinematic chain isomorphism problem. The neural approach provides directly interpretable solutions and does not demand tuning of parameters. We have tested the algorithm by solving problems reported in the recent mechanical literature. Simulation results show the effectiveness of the network that rapidly identifies isomorphic kinematic chains.

[1]  José Muñoz-Pérez,et al.  Design and analysis of maximum Hopfield networks , 2001, IEEE Trans. Neural Networks.

[2]  Wen Miin Hwang,et al.  Computer-aided structural synthesis of planar kinematic chains with simple joints , 1992 .

[3]  S. Merhar,et al.  Letter to the editor , 2005, IEEE Communications Magazine.

[4]  Q. Li,et al.  Some further development on the eigensystem approach for graph isomorphism detection , 2005, J. Frankl. Inst..

[5]  J. P. Cubillo,et al.  Comments on mechanism kinematic chain isomorphism identification using adjacent matrices , 2005 .

[6]  Jiahai Wang,et al.  Stochastic Competitive Hopfield Network and Its Application to Maximum Clique Problem , 2004 .

[7]  Jiahai Wang,et al.  An improved optimal competitive Hopfield network for bipartite subgraph problems , 2004, Neurocomputing.

[8]  A. C. Rao.  An artificial neural network approach to mechanism kinematic chain isomorphism identification , 2002 .

[9]  G. Pawley,et al.  On the stability of the Travelling Salesman Problem algorithm of Hopfield and Tank , 2004, Biological Cybernetics.

[10]  K. H. Hunt,et al.  Kinematic chains for robot hands. I: Orderly number-synthesis , 1995 .

[11]  Enrique Mérida Casermeiro,et al.  An Efficient Multivalued Hopfield Network for the Traveling Salesman Problem , 2001, Neural Processing Letters.

[12]  Yuhu Yang,et al.  A new method to mechanism kinematic chain isomorphism identification , 2002 .

[13]  A. C. Rao,et al.  Application of the hamming number technique to detect isomorphism among kinematic chains and inversions , 1991 .

[14]  Mengkang Peng,et al.  An Investigation into the Improvement of Local Minima of the Hopfield Network , 1996, Neural Networks.

[15]  T. S. Mruthyunjaya,et al.  In quest of a reliable and efficient computational test for detection of isomorphism in kinematic chains , 1987 .

[16]  Y. Takefuji,et al.  An artificial hysteresis binary neuron: a model suppressing the oscillatory behaviors of neural dynamics , 1991, Biological Cybernetics.

[17]  Enrique Mérida Casermeiro,et al.  Modelling competitive Hopfield networks for the maximum clique problem , 2003, Comput. Oper. Res..