基于GIS空间数据模型和多智能体,研究了一种多智能体协作动态路况信息服务系统,通过智能体之间的协作预测路网交通流量在时间和空间的分布,提供更贴近实际状况的诱导路径。在此基础上,基于Swarm平台和Java语言开发实现了实验原型系统,利用实验数据进行实例模拟,对实验结果进行了分析,并对后续的研究工作提出了展望。 Based on GIS spatial data model and multi-agent, a multi-agent cooperative dynamic traffic information service system is studied. Through the collaboration between agents, the distribution of traffic flow in the road network in time and space is predicted to provide more realistic guidance path. On this basis, the experimental prototype system was developed based on the Swarm platform and Java language. The experimental data were used to simulate the experiment. The experimental results were analyzed and the future research work was put forward.