点击次数：

论文类型：期刊论文

发表时间：2012-08-01

发表刊物：Journal of Computational Information Systems

收录刊物：EI、Scopus

卷号：8

期号：15

页面范围：6429-6436

ISSN号：15539105

摘要：An ant colony algorithm model with visual feedback, behavior memory and learning ability is proposed in this paper. Ants can not only perceive the distribution of the target cities around through visual model to improve its searching quality, but also extract experience from the local optimal path to guide the searching activities through memory behaviors and learning models, which could be used to accelerate the convergence rate and to strengthen the searching ability. Then the parallel transformation of the new model are used in the GPU environment. Experiments show that the parallel algorithm under the new model has great improvement in solution quality and solving time compared to other parallel models. While compared with the serial algorithm of the same model, the parallel algorithm could make a good speedup. ? 2012 Binary Information Press.