Research On Data Acquisition System And Positioning Technology Based On Internet Of Vehicles

The integration of intelligent and networked Internet of vehicles is the development trend of the automobile industry,which can not only effectively solve the problem of vehicle management,but also reduce the number of traffic safety accidents and escort safe travel.Efficient and accurate acquisition of vehicle traffic data and positioning information is an important part of the development of Internet of vehicles,and also the basis for the realization of more Internet of vehicles.Widely used by the Global Positioning System(Global Positioning System,GPS)and Inertial Navigation System(Inertial Navigation System,INS)combined with integrated Navigation System to the vehicle position information.However,with the increasing complexity of urban roads,satellite signals will be blocked,multi-path propagation and other problems will lead to the failure of satellite positioning system,while independent INS positioning information will generate cumulative errors over time.Therefore,the thesis focuses on the instability of satellite navigation through theoretical and technical research,and improves the accuracy and stability of navigation from two aspects: terminal equipment and data acquisition,navigation and positioning algorithm.The specific research content is as follows:First,a complete set of data acquisition system for Internet of vehicles is designed and developed,including on-board terminal and server,which can efficiently and accurately realize the collection,upload and storage of vehicle traffic data.The hardware of on-board terminal includes cortex-m3 embedded microprocessor master control unit,positioning data module,OBD module,data communication module and inertia module.The requirements of on-board terminal miniaturization,anti-jitter ability,adaptability to harsh environment and ease of use are well realized,and the hardware and corresponding circuit are analyzed in detail.On-board terminal software uses the embedded RTX real-time operating system to realize remote online upgrade of equipment,satellite positioning information acquisition and OBD interface data acquisition,wireless module to send data and other functions.Secondly,on the server side,the vehicle data receiving,parsing and management procedures are designed and developed to achieve on-board terminals management,upgrade and log view functions.Around the satellite positioning signal failure during vehicle cannot locate problem,in this paper,a sliding window based positioning error modeling ideas,based on the improved XGBoost regression algorithm and predict positioning error compensation,at the same time by PSO to optimize the model parameters of SW-XGBoost,improve the prediction accuracy of the models,improve GPS signals disappear stage vehicle positioning performance of integrated navigation system,realizing a continuous and reliable vehicle in the complex environment of navigation and positioning.Finally,experiments are carried out in real urban roads to verify the effectiveness of the proposed SW-XGBoost algorithm.