车载gps速度数据修正研究及应用车辆工程专业论文.docxVIP

广东工业大学硕士学住论文!!!====暑=====!!!=!!=!=詈=昌皇=!==!竺!!!!=詈皇暑=====!!!!!!!!!詈詈鼍詈!皇!!!!=!=詈===! 广东工业大学硕士学住论文 !!!====暑=====!!!=!!=!=詈=昌皇=!==!竺!!!!=詈皇暑=====!!!!!!!!!詈詈鼍詈!皇!!!!=!=詈===! !!：=暑暑====! AB STRACT Speed is an important parameter of driving vehicle．It not only represents automobile dynamic property,also closely related with the emissions．Global positioning system，GPS，a new method for measuring，has advantages of a11．weather,global，continuous etc．But a number of error sources affect accuracy of measurement，some outliers found in large amounts of GPS data in actual road test．If use the raw data directly．the errors will be transferred by various formulas and models．After delivering，amplifying and accumulating， errors may have serious effect on related researches． The paper is based cooperation project of the Guangzhou Environmental Monitoring Center and Guangdong University of Technology．The existence of outliers in measurement data is proved，all in—depth study the measurement elror causes and related processing method is conducted．An adaptive Current Statistical Model is proposed．and applied to tracking the actual road vehicle．Compare with the data from the CAN bus of the vehicle，the results show that this method Call effectively reduce the random en．or and obtain more reliable data． This paper summarizes research status in the velocity measurement with GPS．Then outline error sources of GPS，introduce common approaches to reduce the GPS measurement error,and explain data post—processing method in detail．According to the knowledge of error analyzing and disposing，and based on the Current Statistical Model。a new model is proposed．The model adjusts maneuvering frequency twice according to innovation vectors． This makes it more suitable to track fast maneuverable target and improve tracking precision． At last，vehicle emission models are established，with speed and acceleration as independent variables．Through comparing actual measured values and model predictions，the validitv of