Title :
Solving traveling salesman problem by simulated electric field method
Author :
Feng, Chun-Bo ; Jiang, Min ; Feng, Jinsong
Author_Institution :
Res. Inst. of Autom., Southeast Univ., Nanjing, China
Abstract :
The simulated electric field method is used to solve the traveling salesman problem. Assume that a unit positive electric charge is placed at each city. A unified static electric field is then formed in the whole territory of the area of cities. The potential differences between each pair of cities form a stereo map. Each city is now considered to be located at a height equal to its relative potential. Imagine that this stereo map is a map of a closed-loop mountain chain which is formed by connecting several conic hills of different heights, sizes and configurations. The traveling salesman problem becomes a problem of optimally programming the conic hills and the closed-loop mountain chain. Then a high-dimension global optimization is reduced to a low-dimension dispersive one, and the computation is greatly reduced
Keywords :
mathematical programming; travelling salesman problems; city; closed-loop mountain chain; high-dimension global optimization; low-dimension dispersive optimization; optimal programming; potential difference; relative potential; simulated electric field method; stereo map; traveling salesman problem solving; unified static electric field; unit positive electric charge; Africa; Automation; Cities and towns; Computational modeling; Computer science; Computer simulation; Dispersion; Joining processes; Polynomials; Traveling salesman problems;
Conference_Titel :
Intelligent Processing Systems, 1997. ICIPS '97. 1997 IEEE International Conference on
Conference_Location :
Beijing
Print_ISBN :
0-7803-4253-4
DOI :
10.1109/ICIPS.1997.669214