Research On Damping And Correction Technology Of Optical Fiber Straptown Inertial Navigation System

The inertial navigation system can calculate the velocity,position and attitude information of the carrier under the initial state by the Newton's second law,which is non-essential to receive or launch any external information to locate and is widely used in various fields because of its autonomy,concealment and security.The accelerometer and gyro of strapdown inertial navigation system are directly installed on the coordinate system of the vehicle.It has become a hot topic of research and application by the simple structure,small volume,low cost and high reliability.Initial error,inertial component error and so on are all important reasons for system errors.If we do not restrain the error items,we will inevitably deviate from the route after long time voyage and cannot locate accurately.From the error model of the system,this paper focus on suppress and compensate the Schuler oscillation error and the divergent position error caused by gyro drift of the optical fiber strapdown system.Therefore,it is very important to add damping and correction links in the optical fiber strapdown inertial navigation system for long voyage.This paper first summarizes the working principle of strapdown inertial navigation system,and introduces the attitude updating,speed updating and location updating mechanism of sins in detail.At the same time,the error equation of the system is derived,and the propagation law of the error equation is discussed in detail and the simulation is carried out.It shows that the random drift will cause the amplitude of Schuler oscillation to increase with time,which proves the necessity of introducing damping link.Therefore,the external horizontal damping of the traditional method based on the classical control theory and the Kalman filter based on modern control theory is introduced in this paper.Subsequently,the error analysis of long time running system is carried out.The drift of the gyroscope is still the cause of the divergence of the position error in a longer navigation system,so the position should be relocated.This paper introduces two points and three points of gyro drift estimation constant calibration,and random comprehensive correction method based on Kalman filtering technique.Then,aiming at the sudden change of the reference speed error of the system in the damping state,it will cause overshoot phenomenon.In this paper,a damping algorithm based on adaptive complementary filtering is proposed.After matching the high speed constant speed network and the phase lag lead network with low frequency characteristics,it is found that the network has large damping ratio attenuation faster,but there will be a greater overshoot when the external reference speed is not accurate.Therefore,the difference between the external reference speed and the system solution speed is added to the adaptive mechanism that selects the damping coefficient in real time to get the optimal damping coefficient which is suitable for different external reference speed errors,and to eliminate the estimated error caused by the sudden change of the environment.Both simulation and vehicle experiments prove that the algorithm can eliminate Schuler oscillation and overshoot.Finally,in view of the changes in the statistical characteristics of the external sensor noise,this paper proposes a correction method based on adaptive Kalman filter and obtains the posterior probability density of unknown measurement noise by variational Bayesian approximation.Compared with the comprehensive correction achieved by standard Kalman filtering,the correction algorithm based on adaptive filtering algorithm can accurately estimate the measurement noise and gyro drift simultaneously when the measurement noise variance of the external device changes.Simulation results,table swaying experiments and sea trial data show that the adaptive filtering algorithm can accurately estimate gyro drift and noise variance,and feed it back to the system.The position error is obviously reduced,and the effect of system resetting is achieved.