Research On Pedestrian 3D Independent Positioning System Based On MEMS Inertial Sensor

With the development of navigation and positioning technology,the demand for pedestrian location information is growing in many applications.At present,most outdoor pedestrian navigation and positioning mainly rely on GPS technology.In the absence of GPS environment,it can only rely on expensive inertial navigation system and the environment in advance of the auxiliary beacon device to achieve pedestrian location information detection,and most of them only satisfy the 2D plane positioning.These are not conducive to promote the application.Therefore,it is a hot topic to study the design of a navigation and positioning system with strong independence,no need of additional basic network facilities and realizing high precision pedestrian tracking in three-dimensional complex environment.In recent years MEMS inertial sensors began to be widely used in the field of consumer electronics.The advantages have small size,light weight,cheap and strong independent detection which make it a widespread concern by researchers.But compared with the traditional inertial sensors,MEMS inertia sensor has these characteristics about lower accuracy,unstable random error and larger algorithm error.Therefore,the solution about the MEMS inertial sensor used in pedestrian 3D positioning technology which has a lot of capacity for development,especially in the sensor error processing and solution process of compensation,these aspects need further study and exploration.In this paper,a pedestrian 3D tracking and positioning system is designed about low cost MEMS inertial sensors.According to the characteristics of multiple sensors,the data fusion algorithm based on the gradient descent method is used to solve the pedestrian gait,which effectively solves the influence of the cumulative error generated in the attitude calculation of the inertial sensor.In order to improve the final positioning accuracy of the system,this paper makes a detailed error analysis of three aspects of pedestrian 3D positioning system based on MEMS inertial sensor,which is integral error,heading error and height error.By using the multi-condition threshold judgement algorithm to improve the accuracy of the step detection,and the velocity integral error compensation is corrected by the zero-speed correction algorithm.The magnetometer and barometer are combined with the inertial measurement unit data.The magnetometer can improve the heuristic heading estimation algorithm,so the pedestrian heading angle correction under different gait can be realized.In view of the long-term data drift of the barometer,the height difference between the inertia system and the barometer is corrected by Kalman filter.Finally,this paper makes a low-cost prototype production and algorithm programming for the whole system.The dynamic error of the system is compared and analyzed in the semi-physical simulation experiment.The three error correction algorithms proposed in this paper are analyzed and evaluated respectively.In a variety of environments,the use of precision of the whole system is experimented and analyzed.It verifies that the system scheme can initially achieve pedestrian tracking in 3D.