Research On The Autonomous Pedestrian Navigation System Based On MIMU

The MIMU-based autonomous pedestrian navigation system(PNS)is a kind of wearable personal navigation equipment.For its low-cost and positioning services in both indoor and outdoor environments,it has a wide application prospect.However,due to the influence of the MIMU's low measurement accuracy,the PNS has a severe accumulated pose error,which greatly limits the application of the PNS.Therefore,the error correction and implementation of the pedestrian navigation system are mainly studied in this dissertation.Firstly,the common coordinate system of PNS and conversion relationship are introduced.The algorithm of pose calculation for PNS is also studied.The pose error model of the pedestrian navigation system is constructed by analyzing the error transfer relationship of PNS,and the filtering algorithm used in the error correction of PNS is also studied.Then the system correction algorithm based on zero velocity and zero angular rate is studied.The motion state of the foot in the pedestrian gait cycle is analyzed,and the system correction algorithm based on the zero velocity is realized by the characteristics that the foot is stationary relative to the ground during the standing stage.The problem of missing detection caused by accelerometer and gyroscope output noise in the gait recognition algorithm is also analyzed.The gait recognition algorithm is improved by the method of noise-reducing filter technique.As the heading drift error can't be suppressed by the zero velocity correction algorithm,the system correction algorithm is realized by combining the zero velocity and zero angular rate information.The height correction algorithm integrating the barometer information and the heading angle correction algorithm integrating the magnetometer information are also studied.The static and dynamic performance of the barometer height information is analyzed.The height correction algorithm based on zero velocity gait window is proposed to solve the problem that the barometer can't be used for system height information correction directly,due to the poor dynamic performance.After analyzing the main interference factors of geomagnetic information,the geomagnetic measurement error model is constructed,and the error compensation algorithm based on ellipsoid fitting is also studied.Then the heading angle correction algorithm integrating geomagnetic information is realized.Finally,the autonomous PNS is realized using the MTi-G-710 sensor and the Raspberry Pi.In this paper,the overall structure of the PNS is introduced with the design and implementation of the hardware and software.Firstly the hardware requirements of the system is analyzed.Then system software was designed and implemented by modularizing software functions.