Research On Road Travel Time Prediction And Visualization

With the development of urbanization and the increasing number of cars,traffic congestion has become an important issue facing the world.Road travel time prediction not only can be used to efficiently manage urban road network,but also help people to arrange travel time and route,energy saving and emission reduction.It is an important research topic in the field of intelligent transportation.Starting from the two main factors of time and weather,this paper analyzes the distribution characteristics of road travel time,uses velocity dynamic relative migration rate to evaluate road congestion and excavate the time-varying law of road traffic conditions and abnormal roads,establishes a time series prediction model of road traffic conditions,and then estimates road travel time.Experimental results on the data set provided by Ali cloud show that the prediction model based on time and weather in this paper has significant prediction effect.Secondly,in order to more intuitively present the road network traffic conditions and travel time prediction results,this paper also carries out a visual study of the vehicle trajectory and road network situation.Especially for the problem that electronic maps are difficult to display a large number of point data efficiently in real time,this paper designs and implements a sampling algorithm and framework for efficiently displaying a large number of point data on an electronic map.Finally,based on the B/S architecture design,a prototype system for road travel time analysis,prediction and visualization is implemented.In summary,this paper establishes a time series prediction model that considers the weather influence coefficient and a visual prototype system for road travel time to solve the problem of urban traffic congestion,and mines abnormal roads.It also solves the problem of efficiently displaying a large amount of point data on an electronic map.