基于多智能体技术的出行诱导管理系统研究

Study on the system of induced travel management based on multi-agent

将实时道路网络动态交通信息与交通管理专家的知识和经验结合起来 ,基于多智能体技术设计了以城市现有交通管理中心为基础的出行诱导管理系统 .该系统通过黑板模型集成了包含在出行诱导管理过程中的交通问题类型确定、分流交通量预测和分流交通量控制 3个智能体 .系统使用一个管理智能体通过黑板模型对这 3个智能体进行集中控制 ,使每一个智能体能够与黑板通信 ,智能体之间在黑板上进行信息分享与协作 .研究表明 ,应用多智能体技术设计的出行诱导管理系统能对出行诱导负效应的产生进行实时跟踪和控制 .

Combining the real-time dynamic network traffic information and traffic managers' knowledge and experiences, the system which is in the existed urban traffic management center is designed for managing induced travel based on multi-agent technology. Agent of deciding the types of traffic problem, agent of forecasting traffic volume diverted and agent of controlling traffic volumes diverted are involved in induced travel management process, and are integrated in this system. A system management agent is used to control these three agents centrally through the blackboard model. Each agent communicates with the blackboard and information sharing and cooperation between agents can be conducted on blackboard. The results show that the system can track and control the occurrence of negative effects of induced travel in real-time.