Urban Road Traffic Travel Time Prediction Algorithm And Software Implementation

Prediction of travel time is one main content of ITS, and it is important to the traffic guidance and route guidance, travel time prediction is able to reflect the traffic situation and trend of road in the future, the accurate travel time prediction can not only improve the decision-making ability of traffic administrative department, and can induce the traveler choose reasonable travel time and travel route and adjust the travel plans, save travel time, improve the efficiency of travel. In this paper, depending on the national science and technology support plan corpus (2011BAH16B01-05)《digital home service mode and new technologies and integrated application research》 research the traffic travel time prediction of urban road and develop the digital home traffic travel information service software under the Android system based on the research of the algorithm, realizing the dynamic shortest path information query before travelling.This paper mainly study the travel time prediction of urban road, including typical sections of city road and intersection sections under signal control as the research object, by contrasting and analysising the basic methods of travel time prediction, Using Kalman filtering algorithm on the travel time data of license plate recognition based on link travel time prediction,and the improving Kalman algorithm, through examples proving that improved Kalman filtering method accelerate the predicted rate of convergence, improving the prediction accuracy, and the predicting precision can well meet the requirements of traffic guidance system. Using traffic wave theory predict the travel time of intersection road under the signal control. First of all, based on the basis of the principle of gathering wave, evanescent wave and the relations between flow rate, density and speed on the road to calculate travel time, Then, using the difference between Link travel time calculation results and the observed travel time data to adjust the section average speed correction calculation results which could be used to train the database of travel time prediction, Finally, by road flow under test match corresponding database to get the corresponding sections of the average speed for travel time prediction, and put the prediction results again for velocity feedback. According to the flow and the error range of travel time after the feedback, put flow and the adjusted average speed into database, in turn, to increase the number of the database sample to improve forecasting model, the experiment data using traffic parameters of the different traffic state of urban road (such as flat peak, peak), comparing with the prediction results with Kalman filter, results show that the traffic wave theory of travel time prediction method is more similar with the trend of traffic flow of urban road, the prediction accuracy is more accurate. Finally, develop transportation information service software under Android System.