Research And Implementation Of Seamless Positioning Based On Multi-sensor Fusion In Smart Phones

With the development of 5G,people's demand for location-based services continues to expand.Accurate and continuous indoor and outdoor positioning has become an indispensable part of people's daily lives.At present,satellite navigation and positioning systems in outdoor environments can already meet people's daily outdoor positioning needs.However,due to the occlusion of buildings and the influence of multipath effects in an indoor environment,the satellite positioning technology cannot be used directly.In order to complete the "last mile" of navigation and positioning,multiple communication technologies such as Wi Fi,ultra-wideband(UWB),and Bluetooth are gradually being applied to indoor positioning,and higher positioning accuracy can be achieved.However,the implementation of common indoor positioning technologies often needs to rely on the deployment of infrastructure and the update of related databases.With the continuous innovation of sensor technology,the pedestrian dead reckoning algorithm(Pedestrian Dead Reckoning,PDR)that does not require external facilities has received extensive attention.Based on the built-in sensors of smart phones,this paper proposes a three-dimensional indoor positioning method based on landmark-assisted PDR and an intelligent continuous switching method between indoor and outdoor to realize the seamless indoor and outdoor positioning of pedestrians based on the existing research foundation and technical difficulties.The specific research content of this thesis mainly includes the following five points:(1)The principle of the PDR algorithm is studied.Aiming at the problem of false peaks or the false troughs in gait detection,a gait detection algorithm based on dynamic threshold is proposed to reduce misjudgments;Aiming at the problem that the magnetometer in the direction sensor is susceptible to large heading estimation errors,a heading estimation method using the Kalman filter to fuse gyroscope,accelerometer and magnetometer is proposed.This algorithm can obtain the user's more accurate positioning information in a short time.(2)Aiming at the cumulative error of the PDR algorithm in the long-term positioning process,this paper proposes a pedestrian dead reckoning technology based on landmark assistance.Considering some known key locations(elevators,stairs)indoors as landmarks.When pedestrians pass by the landmarks,the pedestrian's position is automatically corrected to eliminate PDR cumulative errors.Experimental results show that the improved PDR algorithm can improve the positioning accuracy by 45.28%.(3)In order to reduce the impact of environmental factors on the switching effect of indoor/outdoor environments,an indoor and outdoor environment discrimination algorithm based on the illumination sensor,satellite signal and cellular network signal in the smart phone is proposed.The final test results show that the algorithm can achieve 89.16% accurate detection.(4)Carried out the theoretical research and simulation analysis on the existing air pressure height measurement model,measure the user's moving distance in the vertical direction of the space through simulation calculation,combine the building floor height to estimate the amount of floor change,and combined with the improved PDR algorithm,the indoor three-dimensional positioning is realized.(5)Designed and developed a set of Android-based indoor and outdoor seamless positioning system software,including outdoor positioning module,indoor 3D positioning module based on multi-sensor fusion,and indoor and outdoor environment switching module.The experimental results verify the feasibility and effectiveness of the system.