Research On Indoor Pedestrain Positioning Based On UWB/MEMS Combination

With the highly development urbanization,people spend most of their time and activities in staying indoors.People’s urgent demand for location-based services(LBS)has promoted the rapid development of various indoor positioning and integrated positioning technologies.Because of its complete navigation data types,high data update rate and good positioning continuity,Inertial navigation technology(INS)based on Micro Electro Mechanical System(MEMS)occupies an extremely important position in the field of pedestrian navigation.However,limited by the accuracy of MEMS inertial devices,inertial navigation errors will increase exponentially with time,which cannot meet the needs of long-term navigation.In recent years,the widespread application of Ultra Wide Band(UWB)technology provides a new solution for building high-precision indoor positioning systems.This paper studies the problems of UWB and MEMS in the process of indoor pedestrian positioning,and combines the two to achieve complementary advantages.First,the basic theories in INS and UWB positioning are introduced.Including commonly used coordinate systems,strapdown inertial navigation mechanical layout and error equations,common UWB ranging and positioning models.The advantages and disadvantages of MEMS inertial navigation and UWB positioning systems are analyzed and summarized.Secondly,a passive UWB positioning model is established.Aiming at the positioning error caused by the inaccurate crystal frequency of UWB devices,a crystal oscillator frequency compensation scheme is proposed,and the nonlinear positioning equations of the Newton iteration method are used to solve the problem.The actual measurement results verify the effectiveness of the UWB positioning system established in this paper.Then,in view of the inaccuracy of the threshold-based pedestrian position estimation(Pedestrian Dead Reckoning,PDR)in the step counting process,an improved step-counting scheme is proposed,which reduces the accuracy of the threshold setting.Requires and eliminates the influence of false peaks and false valleys on step counting results.Aiming at the common MEMS indoor positioning process that requires magnetometer assistance,a MEMS indoor relative positioning model is proposed,which avoids the problem of inaccurate initial alignment assisted by magnetometers,and realizes that it can be obtained only by using gyroscope and accelerometer.Relatively accurate indoor relative movement of pedestrians.The effectiveness of the scheme proposed in this paper is verified by simulation and actual measurement.Finally,in view of the problems and limitations of using UWB and MEMS alone in the pedestrian compartment,a PDR-assisted UWB/MEMS combination scheme is proposed.The combined scheme improves the stability and continuity of the indoor positioning results of pedestrians,and makes up for the defect that UWB is interfered by external or human factors and cannot complete positioning,which leads to increased errors of the combined system.The effectiveness of the combined scheme is verified by the measured data.