Sizhe Shen

The performance of integrated navigation using global navigation satellite systems (GNSS) and inertial navigation systems (INS) in challenging urban environments is highly dependent on the quality of GNSS observations, particularly under multipath and non-line-of-sight conditions. Conventional weighting strategies based on predefined empirical relationships with satellite elevation angle or signal quality are limited in characterizing the nonlinear and time-varying characteristics of GNSS measurement errors. To address this issue, this study proposes a deep learning-based adaptive weighting method for tightly coupled GNSS/INS integration. Satellite elevation angle, signal-to-noise ratio (SNR), observation residuals, and normalized observation residuals are jointly utilized to characterize observation quality, and a neural network is employed to estimate satellite-specific measurement noise parameters for constructing the measurement noise covariance matrix. Since ground-truth measurement noise is difficult to obtain directly, the network is optimized using position, velocity, and attitude errors as the training objective. Vehicle-based experiments were conducted in representative environments, including open-sky areas, tree-lined roads, urban street canyons, and overpass sections. The proposed method achieves east, north, and up position root mean square errors (RMSEs) of 0.38, 0.75, and 1.14 m along route A, corresponding to reductions of 15.6%, 17.6%, and 17.4% compared with the combined elevation angle and SNR weighting strategy. Along route B, the corresponding RMSEs are 0.56, 0.71, and 1.18 m, representing reductions of 25.3%, 43.2%, and 34.1%, respectively. Consistent improvements are also achieved in velocity and attitude estimation. These results demonstrate that the proposed adaptive weighting strategy effectively improves the accuracy and stability of tightly coupled GNSS/INS integration under challenging observation conditions.

The quality of Global Navigation Satellite System (GNSS) observations on smartphones is highly susceptible to multipath and non-line-of-sight (NLOS) effects in urban environments, resulting in complex and highly variable observation errors. These challenges highlight the necessity of a reliable stochastic model to ensure robust and unbiased parameter estimation. However, conventional empirical stochastic models, such as elevation-dependent or signal-to-noise ratio (SNR)-based weighting schemes, are often insufficient to capture the rapidly changing stochastic behavior of observations in dense urban environments. To overcome this limitation, an adaptive GNSS stochastic model based on a deep neural network (DNN) is developed by integrating SNR, satellite elevation angle, and post-fit pseudorange residuals, which provide a strong indicator of observation quality and environmental context. Specifically, a fully connected DNN is designed to use SNR, satellite elevation angle, and post-fit pseudorange residual as input features, representing signal strength, satellite geometry, and residual information, respectively, and to learn their nonlinear relationship with measurement uncertainty. The network output is then used to adaptively update the diagonal elements of the measurement noise covariance matrix, thereby realizing epoch-wise adaptive weighting within the Kalman filtering process. The proposed DNN-based stochastic model, together with several conventional models, was evaluated using GNSS observations collected by a low-cost u-blox ZED-F9P receiver (u-blox AG, Thalwil, Switzerland) and a Samsung Galaxy S21+ smartphone (Samsung Electronics Co., Ltd., Suwon, Republic of Korea) during vehicle experiments in dense urban canyons. The code-based single point positioning (SPP) results demonstrate that the DNN-based model consistently outperforms traditional stochastic models under both open-sky and urban conditions. The improvement is particularly pronounced for smartphone observations in severely obstructed environments. The proposed DNN-based model reduces the 3D RMSE from 14.25 m, 13.68 m, and 13.05 m, obtained with the elevation-, SNR-, and integrated elevation–SNR-based models, respectively, to 8.94 m, representing an improvement of approximately 35%. A similar improvement is observed for the u-blox ZED-F9P receiver, where the 3D RMSE decreases from 5.71 m, 4.69 m, and 5.15 m to 3.10 m. These results suggest the effectiveness of the proposed DNN-based stochastic model in mitigating complex observation errors and improving positioning accuracy, providing a promising solution for reliable positioning of low-cost GNSS receivers in challenging urban environments.

With the rapid evolution of Digital Twins and Embodied AI, achieving fast, dense, and high-precision 3D perception in unknown environments has become paramount. However, existing Visual SLAM paradigms face a critical dilemma: geometry-based methods often fail in texture-less areas due to feature scarcity, while learning-based approaches frequently suffer from scale drift and unphysical deformations. To bridge this gap, we propose VGGT-Geo, a novel SLAM system that synergizes generative priors from Large Foundation Models with multi-modal geometric optimization. Distinguishing itself from simple cascaded architectures, we construct a Probabilistic Geometric Fusion framework, consisting of (1) Generative Warm-start, leveraging the holistic scene understanding capabilities of the VGGT, (2) Confidence-Aware Optimization to extract dense features via DINOv3 and predict their confidence map, and (3) a Multi-Modal Constraint Closure that fuses point-line features and metric depth priors to constrain rotational Degrees of Freedom in Manhattan Worlds. We conducted systematic evaluations on TUM, Replica, Tanks and Temples, and a challenging self-collected dataset featuring extreme lighting and texture-less walls. Experimental results demonstrate that VGGT-Geo exhibits superior robustness and accuracy in unseen environments. On our most challenging dataset, it achieves an Absolute Trajectory Error of 4–5 cm and a Relative Rotation Error of 0.79°, outperforming current state-of-the-art methods by approximately 50% in trajectory accuracy. This study validates that synergizing the intuition of Large Foundation Models with geometric rigor is a viable path toward next-generation robust SLAM.

In domains such as autonomous driving, 3D object detection is a key technology for environmental perception. By integrating multimodal information from sensors such as LiDAR and cameras, the detection accuracy can be significantly improved. However, the current multimodal fusion perception framework still suffers from two problems: first, due to the inherent physical limitations of LiDAR detection, the number of point clouds of distant objects is sparse, resulting in small target objects being easily overwhelmed by the background; second, the cross-modal information interaction is insufficient, and the complementarity and correlation between the LiDAR point cloud and the camera image are not fully exploited and utilized. Therefore, we propose a new multimodal detection strategy, Semantic-Enhanced and Temporally Refined Bidirectional BEV Fusion (SETR-Fusion). This method integrates three key components: the Discriminative Semantic Saliency Activation (DSSA) module, the Temporally Consistent Semantic Point Fusion (TCSP) module, and the Bilateral Cross-Attention Fusion (BCAF) module. The DSSA module fully utilizes image semantic features to capture more discriminative foreground and background cues; the TCSP module generates semantic LiDAR points and, after noise filtering, produces a more accurate semantic LiDAR point cloud; and the BCAF module’s cross-attention to camera and LiDAR BEV features in both directions enables strong interaction between the two types of modal information. SETR-Fusion achieves 71.2% mAP and 73.3% NDS values on the nuScenes test set, outperforming several state-of-the-art methods.

The availability of Global Navigation Satellite System (GNSS) raw observations in smartphones has driven research into low-cost GNSS solutions, especially in challenging urban environments, which have garnered significant attention from scholars in recent years. This study proposes an improved smartphone-based velocity-aided positioning method and conducts vehicle-mounted experiments in urban roads representing typical scenarios. The results show that when transitioning from low- to high-multipath environments, the number of visible satellites and carrier phase observations are highly sensitive to environmental factors, with frequent multipath effects. The introduction of robust pre-fit and post-fit residual algorithms has proven to be an effective quality control method. Additionally, using more refined observation models and appropriate parameter estimation algorithms led to a slight 6% improvement in velocity performance. The improved Kalman filter position estimation model (KFSPP-P) strategy, by incorporating velocity uncertainty into the state estimation process, overcomes the limitations of conventional velocity-aided smartphone positioning methods (KFSPP-V) in complex urban environments. In low-multipath environments, the accuracy of the KFSPP-P strategy is comparable to that of KFSPP-V, with an approximate 8% improvement in horizontal accuracy. However, in more challenging environments, such as tree-lined roads and urban environments, the KFSPP-P strategy shows significant improvements, particularly enhancing horizontal positioning accuracy by approximately 50%. These advancements demonstrate the potential of using smartphones to provide reliable positioning services in complex urban environments.