Research On GNSS Signal Propagation Trajectory Error Modeling In Train Dynamic Operation Environment

With the development of train operation control system,moving block relying on accurate and reliable train location is the future evolution direction.Global Navigation Satellite System(GNSS)can provide high-accurate positioning service and reduce the dependence on trackside equipment,thus GNSS-based self-conscious positioning is of great importance to achieve train-borne centralization.However,environment along railway is complex and variable,mountains and tunnels cause the blockage and reflection to GNSS signal propagation,resulting in the train location estimation uncertainty.Research on GNSS signal propagation trajectory error modeling has been a key and urgent issue in railway safety-critical applications.This dissertation proposes a GNSS signal propagation trajectory error modeling and uncertainty assessment method in train dynamic operation environment.The method is based on parametric railway typical environmental feature identification,3D scenario modeling and ray-tracing based GNSS signal propagation modeling to quantify the ground-segment propagation error,combining with the regular space-segment error to build the full-path error model.And the state space model is established to realize the uncertainty assessment in dynamic meansurement process based on the measurement uncertainty theory and the train operation state parameters.The main content is as follows:(1)GNSS signal propagation pseudorange error model is built up and divided into two parts,space-segment and ground-segment.The quantity of space-segment error correction model is calculated and verified with static tests under open scenario.(2)Aiming at the influence of GNSS signal propagation along the ground-segment,this dissertation proposes a ground-segment error modeling method for GNSS positioning under typical railway scenarios.The method identifies typical railway scenarios by parameterized environmental features and hierarchical clustering algorithm,establishes 3D digital scenario models and realizes deterministically GNSS signal propagation model with mirror-image based ray-tracing simulation.Railway typical half-sky and urban canyon scenario are selected as examples to quantify ground-segment propagation pseudorange observation errors.(3)This dissertation studies the train-borne GNSS positioning full-path error model and measurement uncertainty assessment method.GNSS positioning full-path error modeling in railway typical scenarios is realized based on the gaussian mixture model,while train dynamic measurement uncertainty assessment is achieved based on the measurement uncertainty theory and the state space model.Field tests on Beijing-Shenyang Passenger Dedicated Line and sensitivity analysis method is carried out for experiment and verification.The results show that the scenario identification method proposed in this dissertation can accurately divide the environment along the track into five typical railway scenarios.GNSS signal propagation model in half-sky and urban canyon scenarios effectively quantify the ground-segment error,matching the error tendency towards field tests positioning result.Considering the impact of the train surrounding environment on GNSS signal propagation,the full path error modeling and dynamic measurement uncertainty evaluation are realized in different scenarios.The results comparing with field test data proves the effectiveness of the proposed method.Figure 65,table 27,reference 64.