Research On GAT Mangnetic Levitation Gyro Total Station Data Processing And Analysis Based On Kalman Filter

High-accuracy gyro theodolites, total stations, because of the observation time is short,and not affected by weather and other external factors, without complicated astronomicalcalculations, with the autonomy and flexibility advantages are widely used in modernunderground engineering survey and scientific research.GAT mangnetic levitation gyro totalstation developed by the Institute of Survey and Space Information in Chang An Universityand China Aerospace Science and Technology Corporation’16thInstisute is the firsthigh-accuracy mangnetic levitation gyro. It broke through the bottleneck of traditional gyrosuspension-type supporting technology，used magnetic levitation technology to extend thelife of the instrument, and improve the environmental adaptability of the instrument.North-seeking orientation of GAT mangnetic levitation gyro station mainly throughcollecting Stator and rotor current to calculate the north to the declination of gyro axis ofrotation.Therefore the observation accuracy of north to the declination direct affects theaccuracy of the instrument, this also indicates that analysis and processing of the stator androtor current data is the key to improve the accuracy of GAT mangnetic levitation gyro totalstation.Kalman filtering is a highly effective method of dealing with dynamic data. Recentyears, with the expansion of application fields, in order to overcome the traditional Kalmanfiltering method’s shortcomings, gradually robust kalman filitering,fading kalmanfiltering,adaptive kalman filtering etc.This paper will introduce different kalman filteringmethods, summarized in four different Kalman filter algorithm.On the basis of previous research,this paper analysis the characterstic of the stator androtor current data.Based on the dynamic model established by previous researchers throughtime series analysis method, using four different kalman filtering methods to deal with thegyro stator current data. Then analyze the advantages and disadvantages of differentfiltering methods, selecting the best filtering method to deal with more complex rotorcurrent data.To improve the quality of the the Gyro current data,providing a theoretical basisfor further study.