Travel Time Prediction And Route Guidance Of Urban Road Network Under Rainfall Weather

The traffic information guidance system is of great significance for alleviating urban road congestion and improving road service levels.Estimation and prediction of the travel time of the route,which combines the running status of the road and external interference factors,is a very important part of traffic information.With the operation of a large number of floating cars in the road network and the accumulation of vehicle trajectory data,it becomes possible to estimate the real-time travel time of a route from the trajectory data.Between the given start and end points,the travel times of different paths are estimated and predicted,and the path with the shortest travel time is finally selected.However,in the actual route guidance,the selection of the route does not rely solely on the travel time as the sole indicator of selecting the route.When urban roads face different rainfalls,the possibility of accidents on the roads will also increase accordingly.For example,the vehicle traffic is blocked due to water accumulation under the bridge or low-lying sections,and the risks of vehicles passing on the roads are different.Finally,the shortest path Not the best route for vehicles.It is very important for travelers to get accurate travel time and road condition information.This paper uses the fusion analysis of floating vehicle data and rainfall meteorological data running in the city,and finally proposes a traffic risk assessment method based on the road condition information and vehicle attributes of different routes.First of all,the impact of rainfall on the operating conditions of urban roads is analyzed,including road operating levels,four elements of traffic,and slippery road conditions.At the same time,the vehicle anchorage time that may occur when different vehicle chassis heights face different water depths is analyzed.Eventually the road was blocked and impassable.The historical track data of the floating vehicle is used to match the map.In this process,the existing road network topology is interrupted to generate a road network topology based on the smallest road section.The smallest road section is newly combined into an optimal route analysis based on the track of the floating car.Sub-path collection.This method greatly reduces the loss time of computing vehicles waiting for traffic lights at intersections,making the calculation results more accurate and reliable.Considering that the travel time of historical vehicles has a certain dependence on the time dimension,this paper uses a recurrent neural network of Long Short-Term Memory(LSTM)to predict the travel time.In the prediction process,the historical sub-path travel time,the average speed of the sub-path,the starting time of the sub-path,and the current rainfall level are used as input features to predict the results.At the end of this article,the travel risk level of the route is evaluated based on the previously calculated travel time,the safe distance of the vehicle chassis,and the depth of standing water.By comparing the travel time estimation algorithm constructed in this paper with other travel time algorithms,this algorithm has higher prediction speed and reliability.At the same time,after adding traffic risk indicators to route guidance,the vehicle has clear information display and safety guidance in terms of vehicle guidance.