NEURAL-NETWORK IMPLEMENTATION FOR THE OPTIMAL PATH PROBLEM

Authors
AGEEV DA ISTRATOV AY
Citation
Da. Ageev et Ay. Istratov, NEURAL-NETWORK IMPLEMENTATION FOR THE OPTIMAL PATH PROBLEM, Journal of computer & systems sciences international, 37(1), 1998, pp. 118-125
Citations number
22
Categorie Soggetti
Computer Science Theory & Methods","Computer Science Artificial Intelligence","Computer Science Cybernetics","Computer Science Theory & Methods","Computer Science Artificial Intelligence","Computer Science Cybernetics
Journal title
Journal of computer & systems sciences international → ACNP
ISSN journal
10642307
Volume
37
Issue
1
Year of publication
1998
Pages
118 - 125
Database
ISI
SICI code
1064-2307(1998)37:1<118:NIFTOP>2.0.ZU;2-W
Abstract
A new solution to the problem of searching an optimal path in a proble m environment with marked obstacles and obligatory visited points is p roposed. The solution is obtained quickly and independently of the dim ension of the problem. Its algorithmic base is the gradient method ada pted to parallel calculations. A model of neural network implementatio n is presented together with numerical values of tuning parameters. Sp ecific features of the implementation are considered.