Traffic Network Simulation Technology Research

With the rapid development of modern cities, a lot of traffic problems emerged,and this brought us new challenges in traffic research. The critical issue in solovingthese problems lies in two aspects, one is the diffculty in modelling the phenomenonand characteristics of various traffic, and the other is the lack of the traffic simulationplatforms which are capable of efficiently simulating the real traffic network in bothsize and complexity. This paper probes into the fundamental causes of the trafficproblems, and proposes a novel lane-based dual traffic network model. A novel discreterange search algorithm is also presented based on the traffic network model.The existing traffic network model abstract the traffic network into a networkgraph taking the roads as edges and the crosses as vertexs. It’s obvious these modelslose some the key features of real traffic network, such as lane changed, the trafficdiverted, vehicle confluence, vehicle crossing etc.. Thus this paper proposes a new termcalled lane-based dual traffic network which abstract the road and cross into road-laneand virtual-lane respectively. This model can represent any traffic network and rules,and convey the relationship of out and in between roads, road changed, traffic divertedand vehicle confluence accurately.Generally, traditional query algorithms index on the query object for fast query.This paper applys the limited size of traffic simulation object, the locality of the trafficsimulation query, the limitation and continuity of the local transport network, andproposes a Discrete range Nearest Neighbor Query (DrNN) algorithom indexed on thetraffic network. Based on the traditional traffic network model and lane-based dualnetwork model, a three-tier query index structure for the traffic network simulation israised turning the range query problem into a one-dimensional query problem. Thefirst-class is road-level network index, and the second-class is lane-level network index,ande the third-class is discrete lane index structure. With this index structure, DrNNalgorithom can reach a constant time complexity for most queries in traffic simulationwithout losing any simulation accuracy.