Research On The Optimization Of Network-Driven Driving Route Based On Short-term Traffic Speed Forecasting

With the rapid development of China's economy,the rise and growth of the online car-hailing market has greatly met people's travel needs.As a link for drivers and passengers to travel,the online car-hailing platform makes travel transparent and open by providing users with online reservations.However,in the needs of passengers and passengers,there is a large communication conflict between the travel needs of passengers and the driver's driving route,which has also become an important safety hazard that restricts the development of the online car rental industry.For the online carhailing platform,it is one of the issues that the platform urgently needs to solve in order to reduce the contradictions between passengers and passengers and improve the competitiveness of the industry.This paper starts from the perspective of a car-hailing platform,based on the data of the GPS trajectory,order,road network,weather and other data of the second ring network in Chengdu in October 2018.Firstly,through data preprocessing,the time and space distribution characteristics of the data are analyzed to select the appropriate Road research area,on this basis,the average speed of the road is used as the standard to obtain the shortterm speed of multiple time slices per day for each road,and the factors affecting the short-term traffic speed are analyzed to provide a basis for short-term traffic speed prediction;Secondly,under the premise of comprehensively considering the influencing factors of speed,this paper builds the CNN-GRU combined model to predict the shortterm traffic speed of the road.The model uses CNN to extract the spatial characteristics of speed data and GRU to extract the time characteristics of speed data.The experiment proves that the prediction accuracy of the combined model is better than the prediction accuracy of the single GRU and CNN models.this paper aims at the minimum travel cost of passengers,and uses the improved ant colony algorithm to combine the speed prediction results with route optimization and apply it to the actual road area.The empirical results show that the improved ant colony algorithm is superior to the ordinary ant colony algorithm in terms of iteration efficiency and travel cost,indicating that the improved ant colony algorithm can recommend real-time optimal driving routes for drivers.In this paper,the optimal route based on short-term traffic speed prediction for carhailing can save users travel time,reduce travel costs,reduce the possibility of travel delays,and make travel more convenient.For drivers,It can effectively reduce the conflicts with users,reduce the possibility of conflicts with users due to communication,reduce the possibility of being complained,and improve the service level of travel;for online car-hailing companies,it can effectively reduce the contradiction between drivers and passengers,reduce the investment in the cost of disputes between drivers and passengers,enhance user stickiness,enhance industry competitiveness,and increase the public's sense of identity and industry status.