Research On Particle Filter Fusion Positioning Technology In Urban Canyon

With the development of mobile Internet and the popularity of intelligent terminals,people's demand for Location-based Service(LBS)is growing day by day,and accurate location information is the key to achieve LBS.However,in urban canyon,which is one of the most widely used scenarios,the distance between high buildings is short,so that the signals received by Global Navigation Satellite System(GNSS)receivers are blocked,reflected,diffracted and visible satellite distributed stripes by buildings,resulting in positioning error is too large or even can not be positioned.For the above problems,this paper research on the positioning technology in urban canyon,the main work is as follows:Research on improved SM location algorithm based on high score weighting.On the basis of introducing the principle of Shadow Matching(SM)localization algorithm,the 3D building model required by the algorithm is built by using Real-Time Kinematic(RTK)and total station measurement technology.Combined with satellite ephemeris calculate satellites position to predict the visibility of satellites.And finally with the satellite signal to noise ratio(Signal-to-Noise Ratio,SNR)to observe satellite visibility for template matching score to achieve the SM algorithm.By analyzing the satellite SNR observe visibility error of the SM in the real environment,the final positioning result is calculated by the sub-high score candidate position weighted on the basis of the highest score candidate position,so the improved SM localization algorithm based on the high score weighting is proposed.The static experiment results in urban canyon show that the average error of the algorithm is 2.07 m and 0.88 m in along-street and corss-street,which is 30% and 66.7% respectively compared with 2.95 m and 2.64 m of the traditional SM.Which proves the validity of SM positioning algorithm and the superiority of improved SM algorithm based on high score weighting.Improved SM/GPS Fusion Location Algorithm Based on Particle Filter.The basic principles and calculation steps of Kalman Filter(KF),Extended Kalman Filter(EKF),Unscented Kalman Filter(UKF)and Particle Filter(PF)algorithm are introduced.The improved SM positioning algorithm model is established by PF,and the details specific calculation steps is elaborated,but the effect of the algorithm is not obvious.By introducing the GPS velocity information,the improved SM/GPS fusion positioning algorithm model is established by using PF.The dynamic experiment results in urban canyon show that the algorithm has a mean error of 3.57 m in the street direction.Compared with the improved SM algorithm,EKF-improved SM/GPS and UKF-improved SM/GPS algorithm of 5.62 m,4.64 m and 3.74 m,decreased by 36.5%,23.1% and 4.5%,respectively.Research on INS/Magnetometer fusion positioning algorithm based on particle filter.This paper introduces the sensor of XiaoMi 2S intelligent terminals.Based on the analysis of GPS positioning error in the urban canyon,the multi-sensor fusion positioning algorithm of the horizontal velocity of INS and the azimuth angle of magnetometer based on particle filter is realized on the intelligent terminal.The experiment results show that the average error of proposed algorithm is 3.19 m,which is reduced by 76.9% compared to 13.81 m of GPS,and also reduced 34.1% and 33.8% compared to 4.84 m of EKF and 4.82 m of UKF fusion algorithm.Which verifies the validity of multi-sensor fusion positioning on intelligent terminals.