Research On Method Of Real-time Travel Time Collection Based On Simplified Road Network Model

Today, car is an essential tool for people to travel by. When a large number ofvehicles swarm into the road, the traffic congestion is inevitable. So people use pathplanning of intelligent transportation to ease the growing traffic congestion. However,almost all the existing methods of path planning utilize complex road network modelswhich have considered excessive traffic information of roads. Information collection ofthese models are cumbersome and not accurate enough. The recommended path has thepotential to cause the traffic falling into the "secondary congestion”, and cannot satisfythe requirements of travelers.Recently, simplified road network model has been proposed and there are a lot ofresearches and applications of intelligent transportation based on the model, accordingto the application requirements of the city traffic. The simplified road network modelrequires only a few collected road traffic information, but it can indirectly state delaytime, traffic speed, traffic lights, zebra crossings and other complex factors on theeffects of vehicles’ travel time through related calculation, with the features ofcollecting less road information, more simple and practical.Road travel time collection is an important foundation for developing theapplication of city intelligent traffic. This thesis presents some new methods of roadtravel time collection based on the simplified road network model. Main research topicsof this thesis are as follows:①This thesis presents a method of real-time traffic information collection based onGPS terminal, which provides real-time data for the simplified road network model.First, GPS terminal which is similar to the device of the floating car, is designed to dothe work, which used to be done by server, such as GPS signal collection, map matchingand road travel time calculating, and only road travel time is transmitted to the server;Second, the server receives the information, and updates the road network informationtimely by related calculations.②This thesis presents a map matching method with lower complexity, morereal-time and higher matching accuracy, to guarantee the effectiveness and accuracy ofthe data collection. The method, which is based on the weights measure and curvefitting map match ideology, and combined with road network topology information,provides vehicle location information timely and accurately, and can be applied to the vehicle monitoring, real-time traffic information collection and vehicle navigation. Thisthesis also raises the arc-projection algorithm with vehicle position relevant and thedead reckoning algorithm when the GPS signal is invalid.③According to the proposed method, this thesis has developed a more completeprototype system of data collection, and verified the accuracy, efficiency andavailability of the methods above with the repeated tests.In this paper, the method of the real-time travel time collection can providereal-time, accurate and valid road travel time data for city traffic based on the simplifiedroad network model, and has good academic and practical reference value forresearching on city traffic monitoring, vehicle navigation, as well as providing relatedintelligent traffic application system for people.