Research And Implementation Of GPS/INS Integrated Navigation Algorithms

GPS/INS integrated navigation has great application value both in the military andgeomatics field, which is researched both at home and abroad. This paper mainly focused onnavigation algorithms in theory and practice, and some research work has been carried out.The main research work and contributions are as follows:(1) The error equations of the GPS positioning and velocity measurement are derived indetails. GPS position and velocity estimation models based on least squares are given. Sourcesand correction methods of GPS error are analyzed. Measured data from three different GPSreceivers are processed using self-design software, which shows that positioning precision ofthree different GPS receivers is similar, but velocity precision differs a lot; the precision ofrelative velocity measurement is not better than the single point velocity measurement whenusing pseudorange rate observation value; clock bias and clock drift have a significant differentcharacteristics among different receivers.(2) INS navigation equations resolved into the Local Level frame and Earth CenteredEarth Fixed Frame are studied, and the derivation of the INS error equation is greatly focusedon. The derivation of INS error equation based on the ENU position error in the form of angle resolved into the Local Geodetic Frame is a novelty in this paper. The INS error sourcesas well as the initial alignment method are studied. Measured data are processed usingself-design software, which demonstrates that the INS has the characteristics of a short-termhigh precision, but a rapid accumulation of error with navigation time extend.(3) The basic theory of the Extended Kalman Filter for integrated navigation application isstudied. Aiming at the GPS observation abnormalities, a Kalman filter theory based oninnovation inspection is introduced. The Kalman Filter model of three different integratedmodes, loosely coupled mode based on position and velocity, tightly coupled mode based onpseudorange and pseudorange rate measurement, relative tightly coupled mode based ondouble-difference pseudorange and pseudorange rate measurement, are derived resolved intothe Local Geodetic Frame and the Earth Centered Earth Fixed Frame. The correction method ofintegrated navigation lever arm effect is also studied. Measured data are processed usingself-design software under three different integrated modes, and the results are analyzed andcontrasted. The study of the relative tightly coupled mode based on double-differencepseudorange and pseudorange rate measurement is a novelty in this paper.(4)The basic principles of the simulation of trajectory are studied. A trajectory with a leverarm is simulated using self-design software, from which the lever arm effect on integratednavigation is researched. A new Kalman Filter model extended with lever arm parameters based on loose coupled mode is derived. Simulation results show that the model can correctlyestimate the lever arm parameters. The study of lever arm is also a novelty in this paper.