Research On Regional Road Speed Prediction Model And Optimal Route Based On Travel Plan

With the increasing number of cars,the traffic in many cities is worrying,which has a negative impact on social stability and people’s quality of life.Traffic flow forecasting is a key factor in alleviating traffic congestion problems.For traffic flow prediction,many years of research have produced and developed a variety of methods,but there are still some problems.For example,it cannot solve sudden traffic conditions such as traffic accidents,and it cannot cope with rapidly changing traffic conditions.Most of the current road condition prediction methods use historical data sources and real-time data sources.With the development of big data and intelligent transportation,these prediction methods have improved urban traffic congestion to a certain extent.But it is still necessary to predict traffic flow more accurately and efficiently,so as to better relieve traffic congestion.In view of the above problems,this paper proposes to use traveler’s travel plan data to predict future road conditions,and to update and analyze traffic flow information by road sections.Based on the real-time and real travel planning data of travelers,it will be more timely and accurate than previous historical data prediction methods.This paper focuses on the following three aspects:(1)The possibility of data sharing of travel plan is analyzed in detail,and based on the idea of travel plan,firstly,a technical scheme of a future road condition visualization system based on travel plan is introduced;The design of spatio-temporal database based on My SQL is introduced in detail;then,it is further extended on the prototype system framework of our research team,using front-end and back-end development languages,to expand the personalized service function of the travel plan sharing system website and optimize the recommended driving route function,to achieve the optimization of the system website;(2)Taking advantage of the temporal convolutional network to learn temporal features and the adjacency matrix to learn spatial features,a regional road traffic speed prediction model based on TCN is established and improved;The prediction results are displayed,and typical indicators are used to evaluate the prediction results;(3)Design a comprehensive road resistance function based on the traveler’s personal preference;by dividing the time domain,a traffic speed data update method based on travel plan is proposed,and on the basis of the preference model based on comprehensive road resistance,Floyd algorithm in the static determination network is introduced.The algorithm adapts to the dynamic traffic flow and calculates the optimal path for the vehicle to travel.Finally,an experimental closed loop is formed,and the experimental road network of the previous traffic speed prediction model is selected for verification.The path-solving algorithm re-plans the path,resulting in a shorter travel time.Based on travel plan data,this research uses technologies such as big data,cloud computing and deep learning technologies to integrate and process travel plan data resources;Calculate the traffic flow information in the future,such as the number of accompanying vehicles,road traffic status,travel time,etc.,and feed it back to the user;apply the comprehensive road resistance function and the optimal path solution algorithm to provide travelers with a more accurate path planning scheme.Using the future road condition transparency system of travel plan data,the application scenarios of travel plan data are continuously explored,in order to better provide traffic managers and travelers with a method to grasp the road conditions ahead in advance,so as to provide useful reference for solving urban traffic congestion problems.