Reliable and continuous precise GNSS positioning serves as a cornerstone for intelligent transportation, robotic systems, and emerging internet of things (IoT) applications. However, achieving such reliability in urban environments remains challenging due to frequent carrier-phase outliers. Conventional anti-outlier methods are highly dependent on the accuracy of prior absolute pose from INS or other aiding sources. Once the pose estimates degrade, outlier detection may fail or misclassify GNSS measurements, leading to further degradation in positioning results, forming a vicious cycle. To address this problem, this paper proposes a novel carrier-phase outlier-resistance method that is immune to prior pose error. A two-degree-of-freedom TDCP/VDR model is designed to eliminate dependence on prior pose accuracy by leveraging high-precision VDR relative poses and conducting a detailed error analysis of the TDCP and VDR models. Based on this model, we construct a two-step framework that integrates an improved RANSAC approach with a multi-epoch finite state machine (FSM) for reliable outlier detection. Experimental results in complex urban scenarios demonstrate that the proposed method achieves a horizontal position error (CEP95) of approximately 1 m, improving by 63.6% and 56.7% over the IGG-III and Tukey robust methods, respectively, while maintaining the maximum horizontal position error below 2 m. Ablation experiments show that the proposed method maintains strong outlier resistance even with prior absolute position disturbances up to 300 m. Cycle-slip simulations demonstrate high detection recall even when up to 75% of satellites are contaminated by outliers.

In various pedestrian motion scenarios, data-driven pedestrian dead reckoning (PDR) methods have demonstrated strong localization performance, significantly enhancing system adaptability and robustness. However, most existing deep learning models have high computational costs. This makes them difficult to deploy on energy-constrained mobile devices for long-term use. Current lightweight PDR networks adopt a single-input-single-output structure. Although effective in complex dynamics, this design becomes redundant in low-dynamic conditions such as static or slow walking. To address this challenge, we propose a lightweight network architecture named Motion-aware Early-exit Inertial Odometry Network (MEIO-Net). The model adopts a multi-exit design and incorporates a dedicated training scheme. Displacement covariance is used to assess motion difficulty. An early exit mechanism is then introduced to adaptively adjust the inference depth based on dynamic complexity. With a slight improvement in accuracy, MEIO-Net significantly reduces computational cost across multiple datasets. In the self-collected data set, it achieves a 48.9% reduction in FLOPs compared to state-of-the-art lightweight models. In typical simple pedestrian motion scenarios, the computational cost of the proposed model is reduced by 50.5%–80.5%. Ablation studies further confirm that MEIO-Net can flexibly terminate inference according to the intensity of the pedestrian motion. These results demonstrate its strong generalizability and practical potential for real-world deployment. Code and dataset are available at: https://github.com/SunXuehang/MEIO-Net.git.

The pose-only Kalman filter (PO-KF) for visual-inertial odometry (VIO) has demonstrated comparable localization accuracy to optimization-based systems while retaining the computational efficiency of filter-based approaches. However, the performance of point-based PO-KF degrades in textureless environments due to the scarcity of point features, while line features can serve as valuable supplements. To exploit the benefits of pose-only line-feature representation, we extend PO-KF with a pose-only line-feature measurement model and propose PLPO-KF, a unified pose-only representation-based Kalman filter for point-line-based VIO. Specifically, to eliminate reliance on 3D line reconstruction and enable immediate line updates, we develop a trifocal tensor-based pose-only line-feature measurement model with a concise and clear measurement equation. In addition, we introduce an INS-enhanced line optical-flow tracking method to ensure robust and accurate line feature association across multiple frames. Extensive experimental results on both public and private datasets demonstrate that PLPO-KF consistently outperforms state-of-the-art point-based and point-line-based VIO systems. Ablation studies further exhibit the effectiveness of both the proposed line pose-only measurement model and the INS-enhanced line tracking method.

The foot-mounted inertial navigation system (Foot-INS) is a crucial technology for professional pedestrian positioning, unaffected by environmental conditions. However, due to the unobservable nature of the absolute heading, single or dual Foot-INS configurations suffer from significant position drift errors. This paper introduces an innovative pedestrian dead reckoning (PDR) method that combines foot-mounted and waist-mounted IMU with magnetic field vector constraints. Leveraging the fact that the displacements of the foot and waist are consistent when the foot makes ground contact, the proposed method uses the relative displacement estimated by Foot-INS to correct the waist-mounted INS. Building on this, a relative magnetic field vector constraint method is developed using error state clonal Kalman filtering, capitalizing on the similarity of magnetic interference within a local area. The results from 12 tests conducted in typical indoor environments, such as offices and underground parking lots, demonstrate that the proposed method significantly enhances positioning performance in areas with frequent magnetic interference. The positioning error is reduced by more than 49% compared to single or dual Foot-INSs.

Visual-inertial state estimation is widely employed in the Internet of Things, with filter-based visual-inertial odometry (VIO) being a popular algorithm due to its balance between computational efficiency and localization accuracy. However, the localization performance of the commonly used multi-state constraint Kalman filter (MSCKF)-based VIO is suffering from linearization errors in feature three-dimensional (3D) positions and delayed measurement updates. Targeting more accurate and robust localization, we incorporate the pose-only representation into the filter-based VIO and propose a pose-only representation-based Kalman filter (PO-KF) in this paper. Leveraging the decoupling of camera poses and feature positions in the pose-only representation, the proposed PO-KF explicitly eliminates feature 3D coordinates from its measurement equation. As a result, the linearization errors caused by feature positions can be removed efficiently, while immediate updates of visual measurements can be conducted. We also introduce an information matrix-derived base-frame selection algorithm to identify the most suitable base-frames for each feature. Extensive experiments on multiple datasets demonstrate that PO-KF outperforms state-of-the-art VIO systems. Notably, PO-KF achieves nearly a 50% reduction in relative pose errors compared to MSCKF-based VIO. Further experiments demonstrate that PO-KF also exhibits superior robustness while maintaining real-time performance comparable to MSCKF-based VIO.

Accurate and rapid INS state initialization is crucial to ensure the performance of vehicular GNSS/INS integrated navigation. However, in typical urban environments (such as under viaducts and urban canyons), existing GNSS-assisted INS state initialization methods are sensitive to observation outliers. This paper proposes a robust INS state initialization method for vehicle-mounted GNSS/INS integrated navigation. The proposed method first derives the error propagation between the short-term relative navigation (i.e., position, velocity, attitude) and the INS initial state and the GNSS observation error model; then, the high-precision relative pose generated by INS is used to construct constraints between GNSS observation sequences, and the INS state initialization problem is converted into an optimization problem; finally, a two-step optimization strategy is designed to improve the problem of high computational complexity in solving the full-state optimization problem. We use six datasets collected in a typical urban environment to verify the feasibility of the proposed method. The proposed method uses observation sequences within a 10-second to initialize the heading, velocity, and horizontal position with errors of 2.50∘, 0.30 m/s, and 11.1 m, respectively, which are reduced by 73%, 41%, and 14% compared with existing methods.

Location-based services (LBS) have become an integral part of daily life and work for the general public. However, achieving widespread and accurate positioning in typical indoor environments remains a significant challenge, particularly in multi-floor indoor parking lots where radio frequency signals like WiFi are often unavailable. Indoor magnetic matching presents a viable solution, but it requires reducing mapping costs through the use of crowdsourced data. To tackle this issue, we propose an innovative method for constructing magnetic maps using crowdsourced vehicle data. Our approach introduces a multi-user joint vehicle dead reckoning technique based on graph optimization, which provides consistent directional estimates of crowdsourced vehicle trajectories. Subsequently, we establish associations between different vehicle trajectories using multi-attribute features of the magnetic field. Building on this foundation, we propose a global trajectory optimization with inequality and equality constraints to achieve precise estimation of crowdsourced vehicle trajectories. Testing with simulated data from two three-floor underground parking lots demonstrates that the proposed method, utilizing only on-board smartphone sensor data, achieves plane and elevation errors of less than 2.75 meters (95%) and 0.59 meters (95%), respectively. Additionally, the magnetic matching positioning error based on crowdsourced magnetic sequence maps is less than 2.29 meters (95%).

Indoor positioning is an important part of supporting the Internet of Things and location-based services. Crowdsourcing-based magnetic map construction is a key technology to realize wide-area consumer indoor positioning. However, current crowdsourcing-based magnetic map schemes are not suitable for typical indoor scenarios (e.g., shopping malls). The reason is that they ignore the characteristics of crowdsourced data, including short-term trajectory, various pedestrian motion patterns, large-scale data set, and so on. In this article, we propose a novel crowdsourcing-based magnetic map construction method. First, learning-based inertial odometry is used to recover precise user motion trajectories regardless of changes in motion patterns. Then, a keyframe-efficient association method of magnetic time–frequency features is proposed, which is suitable for short-term trajectories of various shapes. Finally, a two-step global estimation optimization is proposed to further eliminate false associations of keyframes and improve the robustness of the method. The feasibility of the proposed method is verified by using a multiuser data set in a typical shopping mall scenario. The proposed method takes a total of 60.8 s to process a 12-h data set (subtrajectories with a duration of 90 s), and the average position error is 1.48 m (with scale correction) and 2.53 m (without scale correction). Compared with the existing crowdsourcing-based magnetic map scheme, the proposed method has been significantly improved in terms of feasibility, accuracy, and efficiency.

Lane-level positioning is a critical technology for supporting assisted driving and autonomous driving applica-tions. However, the Global Navigation Satellite System often falls short in providing reliable positioning (GNSS) due to signal attenuation, obstructions, and multipath in urban areas. Fortunately, typical challenging urban environments, such as tunnels and viaducts, create rich magnetic field features due to abundant ferromagnetic structures, offering an opportunity for magnetic field matching methods to achieve high-precision positioning. This paper presents a novel magnetic field matching/GNSS/Inertial Navigation System (INS) fusion algorithm designed for continuous lane-level positioning in complex environments using cost-effective sensors and computation-saving algorithm. Based on the traditional GNSS/INS tight integration algorithm, this research ensures the performance of the positioning system by enhancing the magnetic field matching and fusion positioning algorithms. First, a coarse-fine magnetic profile matching method is proposed to address the accuracy degradation resulting from the travel distance error of INS-derived trajectory. Second, the magnetic field matching position updates are performed in the vehicle frame, which enables more precise position error modeling. The proposed solution is evaluated through five field tests, covering over 200 kilometers of challenging urban roads. The results demonstrate mean CDF95 position errors of 2.09 m, 1.09 m, and 0.87 m in the forward, lateral, and vertical directions, respectively, and 94.67% accuracy on lane-determination.

The 3-D position estimation of pedestrians is a vital module to build the connections between persons and things. The traditional gait model-based methods cannot fulfill the various motion patterns. And the various data-driven-based inertial odometry solutions focus on the 2-D trajectory estimation on the ground plane, which is not suitable for augmented reality (AR) applications. Tight learned inertial odometry (TLIO) proposed an inertial-based 3-D motion estimator that achieves very low position drift by using the raw inertial measurement unit (IMU) measurements and the displacement prediction coming from a neural network to provide low drift pedestrian dead reckoning. However, TLIO is unsuitable for mobile devices because it is computationally expensive. In this article, a lightweight learned inertial odometry network (LLIO-Net) is designed for mobile devices. By replacing the network in TLIO with the LLIO-Net, the proposed system shows a similar level of accuracy but remarkable efficiency improvement. Specifically, the proposed LLIO algorithm was implemented on mobile devices and compared the computational efficiency with TLIO. The inference efficiency of the proposed system is up to 12 times improved than that of TLIO. Source code can be found onhttps://github.com/i2Nav-WHU/LightweightLearnedInertialOdometergithub.

Global Navigation Satellite System (GNSS) radio occultation (RO) has been widely used in the prediction of weather, climate, and space weather, particularly in the area of tropospheric analyses. However, one of the issues with GNSS RO measurements is that they are interfered with by the signals reflected from the earth's surface. Many RO events are subject to such interfered GNSS measurements, which are considerably difficult to extract from the GNSS RO measurements. To precisely identify interfered RO events, an improved machine learning approach-a gradient descent artificial neural network (ANN)-aided radio-holography method-is proposed in this paper. Since this method is more complex than most other machine learning methods, for improving its efficiency through the reduction in computational time for near-real-time applications, a scale factor and a regularization factor are also adjusted in the ANN approach. This approach was validated using Constellation Observing System for Meteorology, Ionosphere, and Climate/FC-3 atmPhs (level 1b) data during the period of day of year 172-202, 2015, and its detection results were compared with the flag data set provided by Radio Occultation Meteorology Satellite Application Facilities for the performance assessment and validation of the new approach. The results were also compared with those of the support vector machine method for improvement assessment. The comparison results showed that the proposed method can considerably improve both the success rate of GNSS RO reflection detection and the computational efficiency.