| With the development of urban space extending from the ground to the ground and underground,location services need to meet the needs of high precision and high continuity in different indoor and outdoor scenes.At present,the Global Navigation Satellite System(GNSS)cannot meet the continuous and reliable indoor and outdoor seamless positioning services in complex scenarios due to its own vulnerability and limitations.The 5th generation mobile communication technology(5G)cellular network has the technical characteristics of ultra-dense networking,large bandwidth,large-scale array antennas,and can provide wide-area,high-precision observation information such as Time of Arrival(TOA)and Angle of Arrival(AOA).However,due to the characteristics of radio signals,Non-Line-of-Sight(NLOS)errors caused by factors such as occlusion and reflection will also occur.Aiming at the above problems,in order to provide continuous and reliable positioning services in indoor and outdoor seamless environments,this paper studies the model and algorithm of GNSS/SINS combination,5G positioning model and algorithm,and GNSS/5G/SINS elastic fusion navigation algorithm.The main contents include:1.The GNSS PPP/SINS loose and tight combination model and algorithm are studied to achieve outdoor high-precision positioning and provide support for subsequent indoor and outdoor seamless positioning and multi-sensor fusion.Firstly,the error characteristics of the inertial device used in the experiment are analyzed.The Allan standard deviation curve of the device is given by testing,and the error characteristics of the inertial device are obtained by statistics.Then,the loose and tight combination model and algorithm of GNSS PPP/SINS are studied and deduced.Finally,a set of measured data including open environment and complex environment is used to analyze and verify the algorithm.The experimental results show that the CG combination is the best in the GNSS PPP dual system mode.The RMSE of the three-dimensional position is 34%higher than that of the single BDS mode,and the GRCE multi-system mode is 78%and 67%higher than the single BDS mode and the CG dual system mode,respectively.On the whole,the positioning accuracy of GNSS/SINS loose and tight combination algorithm is equivalent to GNSS PPP.When the GNSS is out of lock,the loose and tight combination algorithm can maintain the accuracy in a short time,and then the position error diverges rapidly with time.In this experiment,the positioning accuracy of tight combination is about 11%higher than that of loose combination,and the advantage of tight combination is more obvious in specific scenarios.2.Aiming at the characteristics of 5G system with large-scale array antennas,high carrier frequency and large bandwidth,and more intensive deployment of base stations,the model and method of navigation and positioning using 5G base stations are mainly studied.Firstly,the signal transmission model of 5G CSI-RS is studied from the perspective of time and frequency domain,and the spatial angle and delay information contained in the signal are analyzed.In order to obtain accurate angle and delay information from CIS-RS,a three-dimensional spatio-temporal joint estimation algorithm is proposed,which has high computational efficiency and good estimation accuracy.Then,by analyzing the measurement model of 5G base station positioning and the main errors affecting positioning accuracy.Finally,a 5G base station positioning algorithm based on robust filtering is constructed.The algorithm uses the IGGIII equivalent weight function to adaptively adjust its weight in the filtering process according to the measurement information to achieve the ability to resist abnormal observations.By designing simulation experiments,the effects of 5G new technology and signal-to-noise ratio on TOA,AOA estimation and positioning accuracy are analyzed.The results show that the proposed positioning algorithm based on robust filtering improves the positioning accuracy by 79.1%and 33.4%respectively compared with the weighted least squares and extended Kalman filter,and it has stronger robustness.3.A GNSS/5G/SINS elastic fusion navigation algorithm for indoor and outdoor seamless environment is proposed.The algorithm uses the federated Kalman filter architecture to achieve tight combination in the sub-filters,and does not require the subsystem to solve the position information separately;a flexible fault detection and processing module is designed to suppress the influence of abnormal observation on the whole system.The time update values of the sub-filters are fused by the main filter,and the dynamic information distribution is carried out by considering the quality of the observations,which effectively improves the ability of anomaly detection and suppression.The results of outdoor semi-measured and semi-simulated experiments show that the proposed algorithm can effectively isolate the non-line-of-sight error.Under the condition that the satellite is partially unlocked in the indoor and outdoor transition area,the positioning accuracy of the algorithm is about 57%higher than that of the GNSS/SINS tightly integrated navigation system,and the positioning RMSE reaches 0.37m in the indoor environment. |
PDF Full Text Request