Yidong Lou

Zenith wet delay (ZWD) is one of the major sources of errors in the Global Navigation Satellite System (GNSS) positioning. Existing machine learning (ML)-based ZWD models have limitations in their generalization capabilities and in modeling the high spatiotemporal fluctuations in the atmospheric water vapor. To address these limitations, we introduce an advanced ensemble technique that integrates several base ML models through a Self-Supervised Uncertainty-Based Stacking Ensemble (SUSE). The proposed model enhances the accuracy of modeling and prediction of ZWD for China and its neighboring areas. The results demonstrate that our models, SUSE, outperform other ML, (DL), empirical (GPT3), and traditional models (Askne & Nordius (A&N) and Saastamoinen). In terms of RMSE, our blind model SUSE_A achieves internal and external improvements of 24.36% and 24.24% compared to GPT3. Additionally, SUSE_A outperforms conventional models that use meteorological data (A&N and Saastamoinen) and has comparable performance with other ML and DL models that use meteorological parameters. When surface meteorological data were available, the SUSE_B model reduced the RMSE by 34.48% and 47.89% with respect to A&N and Saastamoinen, respectively. Temporal and spatial analysis reveal the superior accuracy and stability of the SUSE model over other models examined in this research. These results indicate the promising prospects of the SUSE model for GNSS applications.

The ionospheric Total Electron Content (TEC) is a key parameter for characterizing the ionospheric properties. In this study, an efficient multivariable time series prediction model, TEC‐Series Core Fusion and efficient multivariable time series (TEC‐SOFTS), was constructed by combining multiple geomagnetic and solar activity indices. Additionally, the TEC‐SOFTS model was systematically evaluated using CODE Global Ionospheric Maps (GIM) TEC data over the China area from two periods (2020 and 2024) with significantly different solar activity levels. The results show that the TEC‐SOFTS model achieved 1‐hr prediction Root Mean Square Errors (RMSE) of 0.86 and 2.03 TECU in 2020 and 2024, respectively. Compared to the Transformer, Long Short‐Term Memory and NeQuick2 models, the RMSE showed improvements of 0.35/1.02 TECU, 0.83/2.82 TECU and 2.94/9.03 TECU, respectively. The Mean Absolute Errors were 0.64 and 1.53 TECU, showing reductions of 0.29/0.84 TECU, 0.69/2.29 TECU and 2.51/6.95 TECU compared to the aforementioned models. In addition, our research results are verified in two different scenarios: the seasonal analysis of ionospheric TEC during the rising stage of solar activity (2023) and the performance verification under extreme geomagnetic storms during a high year of solar activity (2025). The comprehensive results show that the TEC‐SOFTS model exhibits higher accuracy than the compared models.

As a vital atmospheric component, water vapor influences climate change and drives numerous weather‐related processes. The Global Positioning System (GPS) provides a valuable tool for investigating Precipitable Water Vapor (PWV). This study reanalyzed the raw GPS data from 2000 to 2020 at 25 stations over China to generate a consistent and reliable GPS PWV data set. To obtain homogenized GPS PWV time series, potential artificial shifts in the mean caused by changes in instruments and environmental conditions were identified and corrected based on GPS station log files and comparisons with numerical weather prediction reanalysis PWV from the ECMWF Reanalysis v5 (ERA5) and satellite PWV observations from the Moderate Resolution Imaging Spectrometer (MODIS). A comprehensive assessment of the spatiotemporal variability of PWV is performed using this homogenized GPS data set, while using the ERA5 and the MODIS PWV products for comparison. We first found that both ERA5 and MODIS are slightly wetter than GPS PWV, and ERA5 agrees better with GPS PWV than MODIS does. The annual PWV trends are generally significantly positive across China, and exhibit a west‐to‐east gradient in increasing PWV. The homogenized GPS PWV shows a wetting in eastern and central China during summer and a drying or no trend in winter over central and northern China. The relationships among PWV, specific humidity, and precipitation with temperature more closely follow the scale of the Clausius‐Clapeyron expectation in monsoon‐dominated regions than in the plateau region. This study provides new insights into China's atmospheric moisture dynamics in a changing climate.

Reliable state estimation in railway environments presents significant challenges due to geometric degeneracy resulting from repetitive structural layouts and point cloud sparsity caused by high-speed motion. Conventional LiDAR-based SLAM systems frequently suffer from longitudinal drift and mapping artifacts when operating in such feature-scarce and dynamically complex scenarios. To address these limitations, this paper proposes SERail-SLAM, a robust semantic-enhanced multi-sensor fusion framework that tightly couples LiDAR odometry, inertial pre-integration, and GNSS constraints. Unlike traditional approaches that rely on rigid voxel grids or binary semantic masking, we introduce a Semantic-Enhanced Adaptive Voxel Map. By leveraging eigen-decomposition of local point distributions, this mapping strategy dynamically preserves fine-grained stable structures while compressing redundant planar surfaces, thereby enhancing spatial descriptiveness. Furthermore, to mitigate the impact of environmental noise and segmentation uncertainty, a confidence-aware filtering mechanism is developed. This method utilizes raw segmentation probabilities to adaptively weight input measurements, effectively distinguishing reliable landmarks from clutter. Finally, a category-weighted joint optimization scheme is implemented, where feature associations are constrained by semantic stability priors, ensuring globally consistent localization. Extensive experiments in real-world railway datasets demonstrate that the proposed system achieves superior accuracy and robustness compared to state-of-the-art geometric and semantic SLAM methods.

With the popularization of low-cost receivers, global navigation satellite system (GNSS) provides different types of location-based services (LBS) for various fields such as autonomous driving. High-quality raw observations are the key to achieving high-precision and high-reliability positioning. However, low-cost receivers in urban canyon tend to receive too many non-line-of-sight (NLOS) signals, resulting in excessive GNSS outliers, which greatly degrade the positioning performance. Current NLOS identification methods have limited accuracy and most of the suppression strategies for NLOS are simple elimination or weight reduction strategies, which cannot be effectively applied in complex and changing urban environments. Therefore, this study proposes a method of NLOS identification and suppression assisted by consumer-grade fisheye camera, which improves the traditional Otsu thresholding algorithm by taking into account the characteristics of each RGB channel in sky-view images, and significantly improves the correct rate of NLOS identification. Meanwhile, considering the help of NLOS satellites for satellite geometric configuration, the NLOS satellite with the largest weighted position dilution of precision (WPDOP) contribution is retained and downweighed. Experimental results in urban canyon show that compared with original SPP, the proposed method can effectively prevent the occurrence of positioning anomalies in medium and shallow urban canyon, and can also suppress 49.63% in deep urban canyon, with the horizontal, vertical, and 3D root-mean-square errors (RMSE) improved by 38.24%, 54.39%, and 49.32% on average, respectively. The proposed method is also considerably superior to traditional NLOS suppression strategies in terms of accuracy, continuity, and reliability.

Climate over China is mainly governed by monsoons, which bring frequent and intense convective precipitations. Adequate knowledge of monsoon characteristics is necessary for understanding the development of extreme weather events. Given that the monsoon is usually associated with distinct variations in water vapor between dry and wet seasons and rapid dry‐wet transitions, understanding the distribution and variability of water vapor can help characterize the monsoon. In this work, global navigation satellite system (GNSS) data from over 1,000 stations densely distributed in China were used to obtain zenith tropospheric delays (ZTDs). Combined with ERA5 meteorological data, precipitable water vapor (PWV) grid products were retrieved during the period 2016–2020 to characterize the Asian Summer Monsoon (ASM). To address the issue of data gaps in GNSS‐derived PWV, we employed a random forest machine learning method using ERA5‐derived PWV for data imputation ensuring the integrity of the products. We first utilized the bimodality of PWV statistical distribution and selected a suitable metric to quantify the monsoon impact across different regions. Subsequently, the normalized precipitable water indexes (NPWI) were constructed based on temporal variations in PWV to describe the monsoon movement characteristics. The average monsoon onset in China occurs from late May to mid‐June (Day of year (DOY) 150–170), whereas the monsoon retreat takes place in September (DOY 252–274). The interannual variability of the monsoon onset and retreat times is similar with a variation of around 10 days. Generally, the monsoon advances from southeast to northwest over China with an uneven speed.