Research On Indoor And Outdoor Seamless Navigation Method Of GNSS/UWB/INS Integration

The integrated navigation system is an information integration system for indoor or outdoor navigation.It's an optimization system based on data fusion and processing technology for multisensor and multi-source data.The integrated navigation technology based on Global Navigation Satellite System(GNSS)technology and Inertial Navigation System(INS)technology is widely used in outdoor navigation applications with high precision,high frequency and multiple navigation information.Ultra Wide Band(UWB)positioning technology is a radio navigation system technology with advantages,such as insensitivity to channel fading,low power spectral density of transmitted signals,low interception capability,and low system complexity,et al.It is widely used in high-precision indoor positioning.Centimeter-level precision positioning is achieved under Line of Sight(LOS)conditions.In order to achieve high-precision indoor and outdoor seamless positioning and positioning,this paper studies on the integration theory of GNSS,INS and UWB positioning technologies,including INS fast alignment method,outdoor GNSS/INS integrated navigation model,indoor UWB/INS combination Navigation model,attenuation method of UWB Non-Line of Sight(NLOS),indoor and outdoor GNSS/INS/U WB integration and optimal state estimation method for non-Gaussian distributed random noise.The main work and contributions are recognized as follows:(1)Navigation performance is highly dependent on the initial state parameters of the inertial navigation system.This paper proposes a two-antenna GNSS RTK-assisted INS static initial alignment method using noise covariance adaptive control.The proposed method degrade the initial alignment convergence efficiency of the static base inertial navigation system from the perspective of observation value and filtering strategy:by analyzing the relationship between the baseline vector difference between two antennas and the attitude error obtained by different positioning techniques,the initial alignment of the original INS Zero Velocity Update(ZUPT)measurement model is augmented by attitude error measurement model;GNSS dynamic positioning(RTK)baseline length deviation and RTK baseline solution state estimation error are used for configuration of measurement noise parameter of attitude measurement error equation.The experimental results conclude that the proposed method can effectively improve the alignment efficiency and accuracy.(2)For the GNSS/INS Loosely-Coupled(LC)integration,this paper analysis the disadvantages of the conditiona IAE-based adaptive method,and draws that filter positive feedback and oscillatory estimates would be caused.For better filter stability,accurately identification of the observed anomaly and assurance of the stable INS error calibration,a GNSS quality simulator based on the GNSS positioning accuracy index is designed to tune the measurement noise parameters.The experimental results show that the proposed method can lead better filter stability and accuracy,and better attenuation of oscillation estimation and numerical calculation.(3)For GNSS/INS Tightly-Coupled(TC)integration,the multi-GNSS RTK/INS TC measurement model considering the inter GNSS bias is developed to improve ambiguity resolution.The proposed method considers multiple GNSS systems as a unified system to further facilitate the availability and flexibility of multi-GNSS/INS integrated navigation systems in complex environments.(4)Developing the UWB/INS integration,including the LC integrated model and the TC integrated model.The attenuation method of UWB NLOS error in UWB/INS TC application based on PCA algorithm is proposed.The PCA algorithm for NLOS error suppression based on the UWB/INS tight combination observation model observation data base(sample)is derived.The PC A method is an unsupervised dimensionality reduction method without subjective parameter intervention.The algorithm noise reduction basis mainly depends on the sample characteristics.The experimental results show that the proposed method can significantly reduce the magnitude of NLOS and improve the measurement accuracy.(5)An indoor and outdoor seamless navigation system using GNSS RTK/UWB/INS integrated navigation system is proposed,including LC mode and TC mode.In this paper,the tight combination mode is analyzed.Based on the GNSS/INS and UWB/INS tight combination method proposed in this paper,a multi-GNSS RTK/UWB/INS tight combination navigation method considering system bias is proposed,and combined with the NLOS error proposed in this paper.The experimental results show that the proposed method can improve the accuracy of the integrated navigation system,however,large positioning deviations still exists in the settlement results.The stability of the combined navigation solution needs to be further improved.(6)IAE-based adaptive robust filtering is widely used in INS-assisted integrated navigation applications to fault detection and isolation,or the attenuation of model errors.However,in the seamless positioning environment,personnel random walking,signal occlusion and multipath,IAE can't correctly adjust the noise parameters in the presence of disturbances in both the system noise parameters and the measurement noise parameters.To this end,this paper utilizes Expectation Maximization(EM)-based Adaptive Kalman Filter(EM-AKF),which estimates the state parameters together with the system and measurement noise parameters,and achieves local convergence.The experimental results show that the method can improve the stability and accuracy of the parameter calculation of the integrated navigation system.There are 81 figures,18 tables and 223 references.