Research On Pedestrian Navigation System Based On Consumer Inertial Sensor

With the rapid development of society,people's demand for indoor positioning and navigation is increasing.The positioning and navigation technology based on inertial sensors can realize autonomous navigation without relying on external wireless signals,so it has become a hot research at home and abroad.This paper explores the use of consumer inertial sensor for pedestrian indoor navigation,and studies the indoor pedestrian navigation algorithm to improve pedestrian positioning accuracy.The main research work of this paper is as follows:Firstly,an inertial sensing and measuring unit based on the consumer inertial sensing chip MPU9250 is designed.The unit can measure the nine-axis data(acceleration,angular velocity,magnetic data)of the carrier in real time and send them to the navigation computer through the serial port.Due to its small size and light weight,it meets the hardware requirements of pedestrian navigation system.Secondly,aiming at the gait detection module in the pedestrian navigation algorithm,a method of gait detection using only gyro data is proposed.The large noise and bias error of the gyroscope will cause course divergence over time,so as to reduce the positioning accuracy of pedestrians.However,the traditional zero velocity updating algorithm lacks course observation and can not effectively compensate the error,so the course self-observation method is proposed.The method conducts course observation according to the characteristics of indoor structural architecture and the movement data of feet during walking.In this paper,the Kalman filter in the traditional zero velocity updating algorithm is improved and a comprehensive error compensation method is proposed according to the result of course observation.A fifteen dimensional state variable including the attitude error,velocity error,position error,gyro bias error and accelerometer bias error of the navigation system is established.Observation variable is adjusted adaptively according to the course observation of each step,so as to realize the system error estimation and compensation.Finally,based on the analysis of the movement characteristics of the walking process,The error compensation of the system is explored by using the constraints in the walking.The method of fusing the information of dual feet to improve the navigation accuracy of the system is focused on.Experiments on indoor linear,rectangular and u-shaped routes show that the pedestrian navigation algorithm in this paper can effectively reduce the cumulative error of consumer inertial navigation system,improve course divergence and improve pedestrian positioning and navigation accuracy.