Research On Path Information Management And Data Analysis Method Based On Cloud Platform

With the development of the Internet of Things and the Internet industry,people are gradually entering an era of interconnection of objects and objects.The intelligent monitoring technology for remote devices is gradually developed,and people pay more attention to the operation of devices outside the line of sight.At the government level,state-owned assets belong to the state,but some people with ulterior motives use their positions to facilitate theft of state-owned assets and cause losses to the government.The Government Water Resources Bureau has a task to treat domestic and industrial wastewater,separate the sewage into clean water and sludge,and reuse the sludge.Some drivers and departmental managers use their positions to process sludge in private during sludge transport and dump it in public places to pollute the environment.The project aims to design a system to supervise the sludge transportation process,including vehicle driving monitoring,to supervise the driver's violations,as well as vehicle personnel scheduling control and sludge transportation related information collection and display.Especially in the aspect of vehicle driving monitoring,there are some shortcomings in the traditional inspection methods:(1)In the absence of intelligent prediction.if manual control of sludge transportation vehicles is adopted,the human resources consumption is huge,especially the sludge scheduling tasks are every day.Execution in the middle of the night,the cost of manpower control is large.(2)In the aspect of vehicle behavior prediction,the traditional trajectory prediction algorithm is more difficult in the practical application stage.The accuracy and time complexity of the algorithm are not compatible.It is difficult to realize the requirements of real-time monitoring and early warning,so the feasibility is not high.(3)The modes of various trajectory prediction algorithms are different.The effects are different in different scenarios.When the trajectory prediction is actually designed,it must be designed according to the actual scene.According to the business scenario optimization algorithm,the accuracy of the algorithm is low.To this end,based on the whole vehicle transportation supervision system,this paper proposes a trajectory prediction model for real-time monitoring and warning of vehicles in the process of sludge transportation,and when the driver begins to perform the illegal operation,early warning and early processing will steal the state property.The behavior is contained from the source.The main work of this paper is as follows:(1)Combining the traditional vehicle trajectory prediction solution,this paper analyzes the requirements and design goals in this scenario,analyzes the feasibility of the current prediction algorithm in this scenario,and finds the three-point optimizable space of the current prediction algorithm in this scenario..(2)A trajectory prediction scheme based on variable order Markov model is proposed.Use the idea of integrated learning to train the predictive model.The results of the weak models of each order are combined with different weights to generate a strong model.At the same time.the scheme of the road traj ectory model is designed for the current scene.(3)Combine the model with the whole system to realize a complete vehicle management,monitoring and scheduling system,and simulate and compare the effects of traj ectory prediction.In this paper,the designed model and data storage scheme are compared with the traditional scheme,and the model accuracy is compared.The experimental results show that the proposed scheme has a certain improvement in the prediction accuracy,and the real-time predicted resource consumption is controlled within a controllable range,which is an excellent solution for vehicle behavior monitoring.