The Research Of Algorithm On Indoor Positioning And Navigation Based On MEMS-IMU

The continuous development of modernization has led to an increase in the number of large-scale buildings and an increase in the demand for indoor positioning services.Commercial and personal location services all require the use of accurate indoor location information such as airport halls,supermarkets,and underground parking lots.The rapid development of MEMS makes it very promising for indoor positioning.Inertial devices based on MEMS are used in indoor SINS because of their small size,low cost,and high resonant frequency.The application has been studied extensively.This topic combines an inertial sensor based on MEMS and an electronic compass in an Android system mobile phone,and proposes an indoor positioning navigation algorithm.The main contents are as follows:(1)Summarize and introduce several techniques that have characteristics in the study of indoor positioning.It also analyzes the research methods of universities and research institutes that use inertial navigation technology for indoor positioning at home and abroad.Pointed out its current deficiencies,and confirmed the research direction of the subject.(2)Introduce and analyze the basic principles of pedestrian tracking algorithm and inertial navigation algorithm.The conversion relation between common inertial navigation system coordinate system,carrier coordinate system and navigation coordinate system,several representation methods of attitude,and several acquisition methods of attitude angle are described.(3)Error analysis and initial calibration of the sensors used in the inertial navigation system;By analyzing the characteristics of the collected data,digital lowpass filtering is used to perform initial data processing on the electronic compass data with acceleration data and sliding average filter.(4)According to the problem that the positioning accuracy caused by the difference of the fixed position of each inertial device is reduced,a zero offset correction algorithm is proposed;Based on the analysis of the pedestrian's movement characteristics,the drift error of the inertial device itself is accumulated in the doubleintegral navigation algorithm,which results in inaccurate positioning.An improved zero-speed correction algorithm and frequency-domain integration algorithm are proposed.For the problem of manually setting the threshold for zero-speed correction,a K-MEANS clustering algorithm is proposed to calculate the threshold.Aiming at the problem that the K-MEANS clustering algorithm is very sensitive to the position of the initial cluster center,this paper proposes a method to increase the value between the minimum and maximum values of the respective dimensions of the sample data.For the calculation of the heading angle,the yaw angle obtained by the fusion of the acceleration,the angular velocity,and the magnetometer is integrated with the direction angle obtained by the electronic compass of the mobile phone.(5)Using the atmospheric pressure sensor to obtain the atmospheric pressure data of the pedestrian's altitude and use the formula to convert it into height data to obtain the pedestrian's three-dimensional motion trajectory.Using MATLAB simulation software to simulate the algorithm used to verify the feasibility of using the algorithm and navigation and positioning accuracy.(6)The Android software development environment was configured to build an indoor positioning and navigation system based on inertial devices and Android smart phones.The real-time display of data acquisition,storage and processing,motion trajectory screens and pedestrian position height is realized,realizing the pedestrian real-time navigation function in the indoor environment.