基于网络服务模式的动态路径规划蚁群算法

随着云计算、移动互联网等IT技术的发展，通过网络提供动态路径规划服务能够进一步改善人们的出行质量。网络服务模式下的动态路径规划要求系统能够同时为多用户提供最优路径。在多态蚁群算法基础上，借鉴最大最小蚂蚁系统及自然界优胜劣汰思想，考虑共享侦察蚁群得到的初始道路信息素，提出两阶段蚁群算法。实验结果表明，两阶段蚁群算法不仅在收敛速度上有所提高，且适应网络服务模式下的多用户实时导航需求。

Ant colony algorithm for dynamic path planning based on network service mode

Along with the development of the IT technologies such as cloud computing and mobile Internet, people''s journey quality can be improved by providing the network-based dynamic path planning service. Dynamic path planning based on the network service mode should be able to provide the shortest path for multiple users simultaneously. A two-stage ant-algorithm is proposed in this paper. It is based on polymorphic ant colony algorithm, refers to the max-min ant-system and the survival of the fittest ideology, and considers sharing of the initial path pheromone which is generated by the reconnaissance ant colony. Experimental results show that the two stage ant colony algorithm improves the speed of convergence. It is adapted to the concurrent requirements of real time navigation in network service model.