Research Of Spatiotemporal Registration Technology For Integrated Navigation System

In essence,the integrated navigation system is a system that comprehensively processes and optimizes the navigation information surveyed by each navigation device.Therefore,multi-source information fusion technology is the key technology of the integrated navigation system.Multi-source information fusion,also known as multi-sensor information fusion,can more accurately estimate position and velocity information by comprehensively collating the measurement information of multiple sensors.In the actual multi-sensor system,the measurement data of each sensor is not synchronized due to the different execution tasks of the sensors,the advantages and disadvantages of their own performances,and the differences in the environment.Therefore,they cannot be directly sent to the fusion center.For the fusion processing,it is necessary to convert the measurement data of the sensor in different spaces at different times to a unified fusion time and space,that is,perform space-time registration processing.In order to make the integrated navigation system obtain more accurate and reliable navigation information,this paper decided to study the time registration algorithm and spatial registration algorithm of the integrated navigation system.Firstly,the paper analyzes the basic problems of space-time registration of integrated navigation system from four aspects: registration task,time and space error source and influence,space-time registration method and registration performance evaluation criteria.It is clear that time registration in multi-sensor integrated navigation system mainly solves the problem that the sampling period and starting time of the sensor are different.The spatial registration mainly solves the problem of coordinate system and sensor spatial deviation correction.Secondly,the problem of time error caused by different sampling period of the sensor is solved by interpolation extrapolation method.Considering that the motion model of interpolation extrapolation method is too simple,the interpolation extrapolation method is in linear shifting motion and curve.Under the variable speed motion model,the algorithm formula is improved.Considering the existence of the measurement error of the sensor sampling data,the influence of the measurement error of the sampling data on the registration accuracy is analyzed.To reduce the sampling data measurement error,according to Karl The data filtering function of the MANN filter is based on Kalman filter for time registration to improve the registration accuracy.The validity of the registration algorithm is verified by simulation.Thirdly,theoretical analysis is carried out on the influence of the synchronization of the sensor on the registration accuracy,and the time registration problem is solved by extending the Kalman filter to estimate the time difference of the sensor,so as to obtain the time difference after eliminating the startup time.The data estimation value of the sensor;considering the unknown statistical characteristics of the extended Kalman filter measurement noise,the filtering accuracy may be degraded or even diverged.For the problem,the online estimation of the measurement noise is performed to adapt the extended Kalman filter.Improvement,verifying the registration algorithm by simulation and improving the effectiveness of the filtering algorithm.Finally,the problem of coordinate system in navigation is described mathematically.Based on the principle of lever arm effect in inertial navigation system,the principle of lever arm generation between multi-sensors in integrated navigation is analyzed and the speed error caused by lever arm effect is derived.Aiming at the influence of the multi-sensor inter-arm arm effect on the speed estimation in the navigation system,the speed error caused by the lever arm effect is compensated by the lever arm error compensation model,and the simulation verification is carried out.