Research And Implementation Of Pedestrian Seamless Navigation And Positioning System Based On Smartphone

In recent years,the urbanization process of China has been accelerating.With the introduction of some newly arisen concepts such as smart city and intelligent transportation,the demands of location based services from individuals and markets are growing.With the continuous innovation of microelectronics technology,the performance of smartphone’s chips increases.Built-in MEMS sensors in smartphone become richer and richer,smartphone becomes a reliable platform of pedestrian navigation.The pedestrian navigation positioning technology based on smartphone has become an important branch of the navigation field.The research and implementation of the indoor/outdoor seamless navigation and positioning system,which is based on the smartphone,has been progressed in this paper.Firstly,the present domestic and overseas condition of the pedestrian positioning technology is introduced.On the basis of the foundation and technical difficulties of existing research,the indoor and outdoor seamless navigation algorithm based on smartphone has been proposed.Optimized AdaBoost algorithm based on WIFI signals are used to distinguish the environment which is inside or outside the building,so that the corresponding indoor or outdoor positioning method can be selected.The seamless switch of indoor and outdoor positioning mode has also been designed.As the satellite signals cannot be received indoors,UWB and WIFI are introduced to assist inertial sensors.To solve the problem of the quick change of the zero offset and the error,the characteristics of the low cost inertial sensors,which are built in smartphone,are analyzed,and the algorithm of optimizing pedestrian dead reckoning has been put forward.Compared with the traditional strapdown inertial navigation algorithm and the magnetic heading assisted algorithm,this algorithm suppresses the rapid divergence of the positioning results and solves the problem of indoor pedestrian positioning in a short time.In order to achieve the high precision indoor navigation with limited cost,the extended Kalman filter is introduced into the integrated navigation algorithm of UWB and PDR so that the corrections of PDR’s heading and position are accomplished.Considering the demand of indoor environment such as shopping malls,the K nearest neighbor algorithm based on the WIFI,which is optimized by using heading information,are studied to meet the needs of the pedestrian’s indoor positioning.In view of the fact that the WIFI signals are susceptible to environment influence,the unscented Kalman filtering algorithm is adopted to realize the correction and smoothing of the WIFI fingerprint location result with the help of PDR.In order to distinguish the indoor and outdoor environment exactly,the optimized AdaBoost algorithm in the field of machine learning are studied by using WIFI signals.It realizes the intelligent selection of indoor and outdoor navigation mode,which can improve the performance of whole positioning effectively.Finally,the Android application of pedestrian seamless navigation is developed on the eclipse platform,realizing the indoor positioning based on the inertial sensor in smartphone,the optimization positioning based on WIFI and the indoor and outdoor discriminant method based on the optimized AdaBoost algorithm.The effectiveness of the algorithm and the feasibility of engineering are proved by the experiment of indoor and outdoor long time walking.