Research On Key Technology Of Traffic Information Service Composition Based On AI Planning

With the rapid development of Internet technology and the intensive study of vehicular networking,traffic information system is becoming more and more intelligent.However,facing more complex traffic environment and the demand of people traveling convenient,the simple and single service cannot supply all travelers.How to compose a number of single-function traffic information services through a given standard to generate a feature-rich service schema,becomes a hot issue in the intelligent transportation.By analysis of the characteristics of traffic environment,the description framework of traffic information service was constructed,and further,the AI plan-based composition method was provided in this paper.The main work and innovation of this paper can be summarized as follows:1.The common Web service description and the hardware and software resource under traffic environment were discussed.Extending the OWL-S to characterize the traffic services,the RS-OWLS was given to describe the traffic information service.The RS-OWLS was modeled and the knowledge ontology library in the transportation domain was constructed and by Protégé.2.Considering AI planning,the composition method for traffic service based on HTN planning was provided.Under the generated sequences for the future service composition,a fine-grained matching algorithm was proposed to obtain an executable composition scheme.3.The characteristics of traffic information service system and the advantages of agent management were analyzed.Based on this,a task-aware service composition framework was illustrated using a special travel situation.Experiments shown the adaptability and feasibility of the proposed framework.