| With the acceleration of urbanization in China,the number of urban motor vehicles has sharply increased,leading to increasingly serious problems such as traffic congestion and safety.Intelligent transportation systems utilize advanced scientific technologies such as computers,electronics,and communication to effectively solve transportation problems.Among them,traffic prediction and path planning are two important research directions.Traffic prediction and path planning are two important research directions.Traffic prediction is the process of analyzing a large amount of historical traffic data to predict future congestion conditions and provide reference for traffic system control.Path planning mainly involves planning the optimal path from the starting point to the destination point based on the needs of travelers and the real-time changing traffic patterns of the road network,in order to achieve the maximum utilization of the entire road network resources.Therefore,this article achieves traffic prediction based on multi attention and plans a more efficient path based on the predicted future road conditions.The main work of this article is as follows.Firstly,this article proposes a traffic flow prediction model based on multi attention,and conducts in-depth analysis of the spatiotemporal correlation of traffic flow.In view of the problems of the previous model that cannot fully summarize the structural characteristics of the road network and the output accuracy,this model uses the method of bidirectional diffusion convolution to simulate the directionality of the road,and then encodes and distinguishes different time periods by hot spot coding,combines attention with the above information to accurately extract the feature information,and finally uses the gating mechanism to control the transmission of information and output calibration results.Compared with other prediction models,experimental results show that the proposed model is superior to other models,which further improves the prediction accuracy.Secondly,in the existing path planning algorithm,static path planning is difficult to adapt to the changing road network,easy to enter the congested road section,dynamic path planning needs to periodically monitor the road conditions and re-plan the route,so as to avoid congestion,but it will cause the planned path to change repeatedly when the road conditions fluctuate greatly,making it difficult to improve the traffic efficiency.In response to the above issues,this article utilizes the foresight of traffic flow prediction to capture future information on the road network,identify more promising road sections,shorten vehicle travel time,and thereby improve the traffic efficiency of the road network.Finally,this article designs and implements a travel information service system,which embeds the traffic flow prediction model and path planning algorithm.For the current urban road network,the system server analyzes and processes the collected data,trains a practical traffic flow prediction model,predicts the traffic flow and change trend of the road,and uses the path planning algorithm proposed in this article to avoid congested sections of urban roads and plan the path reasonably.In summary,this article combines traffic prediction and path planning techniques to propose a new traffic flow prediction model,which can provide more accurate prediction results and use the prediction results for path planning,effectively shortening vehicle travel time and improving the traffic efficiency of urban roads.Finally,integrate them into the system platform to provide services for users. |
