Design and Simulation of SINS/DVL Integrated Navigation System Based on CKF
|School||Harbin Engineering University|
|Keywords||Strapdown Inertial Navigation System Integrated Navigation System Doppler Velocity Log Extended Kalman Filtering Cubature Kalman Filtering|
With the further deep going of the Marine research，the requirements of the integratednavigation system adopting in the vehicles which work under or over the water weregrowing harsher. In this paper, an integrated navigation system works under or on the wateris studied, which based on Strapdown Inertial Navigation System (SINS) and DopplerVelocity Log (DVL), this integrated navigation system integrated SINS calculating speedand DVL real-time, high precision velocity information, using kalman filter technology torestrain the error of SINS accumulation over time. And due to the low precision of extendedkalman filtering, apply Cubature kalman filtering (CKF) technique to solve nonlinearSINS/DVL integrated navigation initial alignment and navigation problem.Firstly, in this paper, the main inertial coordinate system and basic equation of inertialnavigation systems is introduced, the working principle, error sources, nonlinear errormodel derivation and mathematics of strapdown inertial navigation systems are discussedhere. According to the doppler effect, single beam, double beam, four beam dopplervelocity log measuring principle were deduced, and analyzed the advantage of four beamdoppler velocity log.Secondly, this paper analyzes the error model, error equation, the basic principle andcombination modes of SINS/DVL integrated navigation system. Kalman filteringmechanism and filtering process were introduced, in order to further solve the problem ofnonlinear filtering, the nonlinear state equation and measurement equation based onextended kalman filtering were established. And simulations of the initial alignment,uniform linear motion, uniform circular motion in nonlinear SINS/DVL integratednavigation system were completed. Simulation results showed that EKF can reduce theinitial alignment error and improve the precision of navigational positioning to a certainextent.Finally, cubature kalman filter technology was introduced in detail, analyzed thespherical-radial cubature rule and CKF filter algorithm was deduced, the nonlinear stateequation and measurement equation based on cubature filtering were established, andcompared with EKF filtering results. Simulation results showed that under the samesimulation conditions and initial state, CKF had higher precision than the EKF filter, is a kindof better than EKF nonlinear filter. Not only can greatly improve the precision of initial alignment, effectively solve the SINS initial alignment problem under a large misalignmentangle, and in navigation stage filtering precision is still higher than that of EKF. In the end,adapt the real navigational data to validate, the results confirmed the CKF effectiveness inpractical navigation system.