Research On Receiver Autonomous Integrity Monitoring Algorithm In Urban Complex Environment

With the wide application of Global Navigation Satellite System(GNSS)and the rapid development of smart cities,the demand for positioning in urban environments is becoming more and more obvious,and the effectiveness of positioning is further improved.At present,in many urban positioning scenarios,the main problem that users face is no longer the inability to obtain a sufficient number of visible satellites,but how to correctly detect and exclude those signals that are severely affected by the non-line-of-sight and multipath errors,so as to ensure a reliable positioning result.This dissertation takes the complex urban environment as the background,and studies the related technologies of Receiver Autonomous Integrity Monitoring(RAIM).The main contributions are as follows:(1)Aiming at the problem that the occlusion environment on both sides of urban roads affects the performance of vehicle navigation and positioning,an optimized weighted RAIM algorithm based on vehicle heading angle is proposed.Combined with the characteristics of urban roads,this algorithm introduces a scale factor into the classic high angle-to-carrier noise ratio weighted model.The determination of the scale factor depends on the relationship between the moving direction of the vehicle and the position of the satellite,which makes the new method more suitable for the urban environment with serious occlusion on both sides.Simulation results show that compared with the existing classical optimal weighting model,the 3D positioning accuracy of the weighting algorithm in urban environment is improved by about 9.1%,and the ability of satellite fault detection and recognition is also significantly improved.(2)Aiming at the problem that the observations of the receiver are more likely to be polluted in the complex urban environment,a multi-fault RAIM algorithm based on Random Sample Consensus(RANSAC)is improved.By approximately estimating the user equivalent ranging error of the preferred constellation and introducing the Position Dilution of Precision of the satellite constellation,the evaluation function of classic RANSAC is improved and optimized.Considering various factors affecting the positioning accuracy,a more effective consistency detection method is proposed.The simulation results show that compared with the RAIM algorithm and the classic RANSAC algorithm,the optimized RANSAC algorithm has a considerable improvement in the positioning accuracy in the urban environment,especially in the case of multiple faulty satellites,the improvement effect of the optimized RANSAC algorithm is more obvious.(3)The detection threshold of traditional RAIM needs to be manually determined based on experience and debugging,which limits its performance and scene adaptability.To solve this problem,this thesis proposes a new thresholdless RAIM algorithm based on marginalization probability.Different from the traditional RAIM,the new method uses the marginalization probability theory to marginalize the measurement noise via discrete sampling,and the marginalized probability which represents the confidence of measurement is obtained.Further,the optimized weights are constructed to suppress the measurement errors adaptively,which solves the problem of estimating measurement noise and setting detection threshold artificially in traditional RAIM algorithm.To verify the feasibility and performance of the method,GNSS positioning experiments were carried out in different scenarios.The results show that in static scenes,compared with the optimally configured RAIM algorithm,the 3D positioning accuracy of the new method improves by about 15.7% and 20.1% in the single GNSS and multi-GNSS cases,respectively.In dynamic scenarios,the new method improves the 3D positioning accuracy by about 7.6%and 8.6% in the single GNSS and multi-GNSS cases,respectively.