A New Ant Colony Algorithm Based on Dynamic Local Search for TSP
| Autor: | Jie Luo, Shulun Zhou, Ling Huo, Haisheng Qin |
|---|---|
| Rok vydání: | 2015 |
| Předmět: |
Mathematical optimization
business.industry Computer science Ant colony optimization algorithms MathematicsofComputing_NUMERICALANALYSIS ComputingMethodologies_ARTIFICIALINTELLIGENCE Travelling salesman problem Tabu search Guided Local Search Local search (optimization) business Hill climbing Metaheuristic Premature convergence |
| Zdroj: | 2015 Fifth International Conference on Communication Systems and Network Technologies. |
| DOI: | 10.1109/csnt.2015.241 |
| Popis: | As the traditional ant colony algorithm(ACA) for solving TSP(traveling salesman problem) is easy to fall into a local optimal solution and slow convergence, also the quality of solution is not ideal. An ant colony algorithm based on dynamic local (DLACA) search is proposed, means that each ant has the ability of local search and it can use the ability according to the real-time condition, which enhances the algorithm's search quality and improve the stabilization of solutions; meanwhile, used dynamic policy to updated pheromone. After each travelling, if find a better road, this better road is allowed to update the pheromone severely, which prevents premature convergence. Besides, the combination of dynamic local search and local optimal jumping can again to the stagnation of the algorithm. The traditional ACA and DLACA are used to solve TSP are simulated by Matlab, and the results show that DLACA algorithm can obtain the known optimal solution within the stipulated time as well as the stabilization of solution is also better. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|