Inertial Navigation System Based On Multi-source Information Fusion

Inertial navigation systems can achieve autonomous positioning in indoor and outdoor complex environments.However,due to its error divergence,the accuracy of positioning technology centered on inertial navigation systems does not meet social needs.Therefore,the thesis uses multi-source information fusion technology to integrate pedestrian footsteps,geomagnetism,and maps on the basis of inertial navigation,and to achieve high-precision positioning of inertial navigation with foot-pedestrian trajectory estimation.The thesis introduces the basic working principle of the inertial navigation system and the principle of inertial device error model,and simulates and analyzes the error characteristics of the MEMS inertial device;Analyzing the characteristics of divergence based on MEMS navigation system caused by integral estimation;Summarizing the algorithm flow of Kalman filter,Extended Kalman filter and Particle filter related to multi-source information fusion technology.The thesis studies the method of pedestrian trajectory estimation based on inertial navigation system;By comparing different zero-speed detection schemes,a zero-speed update method for suppressing error divergence in pedestrian trajectory estimation techniques is used to improve and verify the differential zero speed detection scheme;By analyzing the observation equations and state equations of the shoe-bound pedestrian trajectory estimation system,the Kalman filter based on the error state is used to realize the pedestrian trajectory estimation system based on pure inertial navigation;By analyzing the changes of the system’s heading error estimation and introducing asymptotic attenuation factor,the pedestrian trajectory estimation system is improved.Finally,the thesis combines the intrinsic shortcomings of pure inertial navigation,fusion of geomagnetic information and data map information to improve it.The thesis analyzes the geomagnetic conditions in indoor and outdoor as well as pedestrian behavior information,proposes an effective heading deviation estimation scheme,and uses an extended Kalman filter to achieve an indoor and outdoor compatible geomagnetically integrated inertial navigation positioning scheme;By enhancing the particle randomness of particle filtering and improving the particle resampling method based on pedestrian trajectory information,a robust robust map fusion inertial navigation scheme is realized on the basis of geomagnetic fusion;Comprehensive simulation of inertial navigation system based on multi-source information fusion and comparison of its positioning accuracy.