Modeling And Optimization Of Dynamic Route Guidance System Under Vehicular Ad-Hoc Networks

With the speeding up of urbanization and the development of automobile industry, the contradiction between the capacity of existing urban road network and the continually growing traffic demands becomes more and more acute, which leads to increasingly traffic congestions. Facing to these traffic problems around the world, application of Intelligent Transportation System (ITS) can effectively control and induce the traffic flow so as to alleviate traffic congestion and improve the traffic environment.Vehicular Ad-Hoc Networks (VANET), which is a new research issue in ITS, attracts extensive attention gradually. Most existing traffic management systems utilize traffic control center to process the traffic information, and vehicles are only information receivers. While, vehicles in VANET can communicate with each other, so that they can acquire real-time traffic information to generate and change their route dynamically. Therefore, VANET effectively improves the driving efficiency.Based on complex large-scale urban road network, the thesis utilizes advanced theories and methods such as Prospect Theory, Game Theory, aad Bee Colony Optimization Strategy to model and optimize the Dynamic Route Guidance System (DRGS) under VANET environment. A comprehensive and efficient DRGS is established by a series of means such as traffic guidance zone partition, route searching district limitation, route choice model establishment, and route searching optimization. The main research works are as follows:The thesis firstly discusses the purpose and meaning of the issue, and then introduces the research status both at home and abroad in VANET and DRGS. After the analysis of the application prospect and value of VANET and problems in DRGS, the thesis proposes a series of models and algorithms in DRGS.To deal with the complexity, time variance, and uncertainty of traffic condition, a route choice model based on Prospect Theory is proposed. The proposed model takes the incomplete rationality of traffic behavior led by the traveler’s subjective initiative into account. The introduction of PT to DRGS can accurately describe the decision-making process under uncertainty. The proposed Prospect Theory-based model gives a more realistic and reliable route that individual usually takes, and overcomes the shortcomings of EUT in the description of route choice behavior.For the Dijkstra algorithm is difficult to meet the real-time requirement of the dynamic route guidance system, a threshold-based restricted searching area algorithm is proposed. This algorithm, applied to the complicated urban road network model, realizes the optimal route search in the dynamic route guidance system. On basis of the classical Dijkstra algorithm, this algorithm introduces a threshold-based restricted searching area mechanism. The proposal uses the spatial distribution feature of the real road network to restrict the searching area by setting up a reasonable threshold value, which reduces its searching size, so as to enhance its efficiency. The proposed algorithm is faster than traditional algorithms in large-scale urban road network, and it shows a good performance in the dynamic route guidance system.For the existing problems such as computation complexity and poor flexibility of traditional traffic zone partition, which cannot reflect the characteristic of traffic flow, a Shapley value-based traffic zone game partition algorithm is proposed in order to provide reasonable and effective traffic zones in dynamic route guidance system. Firstly, the algorithm confirms the traffic zone core by using similarity theory with road section as basic computation unit. Then, the traffic zone partition is abstracted as the game process. Taking the Shapley value as the game convergence criterion, the proposal realizes the traffic zone partition accurately and reliably. The proposed algorithm can also decrease the complexity of vehicle routing in Dynamic Route Guidance System to improve its real-time performance.On basis of proposed traffic guidance zone partition, a bee inspired zonal vehicle routing algorithm is proposed to provide a reasonable and effective optimal route for the Dynamic Route Guidance System. The proposal simulated the bee foraging phenomenon in the biological system to synchronously compute the optimal routes in each traffic guidance zone. Simulation results show that the algorithm has higher computation efficiency under the precondition of providing the global optimal route.The traffic behaviors of commuters bring traffic congestion in peak hours. The research of commuting behavior will remit the traffic congestion during peak hours. According to the commuting characteristics, a route choice model based on Game Theory is proposed to provide reliable route choice to commuters in real and imperfect situation for commuters. The proposed model treats the alternative routes as game players, and utilizes the precision of predicted information and familiarity of traffic condition to build a game. It can provide the optimal route by solving the Nash Equilibrium of the game. The proposed model can not only describe the commuters’routine route choice decision exactly, but also provide reliable route choice to commuters.Finally, a dynamic route guidance system is established based on middleware. The system can both publish real-time traffic conditions and provide optimal routes. Meanwhile, the system is also a simulation environment for the proposed models and algorithms in the thesis. It provides a powerful technical support to verify the validity of the proposed models and algorithms.Based on the above research results, the thesis concludes the entirety and prospects the further research directions in the end.