A new model improves GPS Block IIF satellite yaw during eclipse-season maneuvers, reducing phase-residual signatures and producing smoother clock estimates. The revised model captures actual Block IIF eclipse-season behavior, supporting more consistent precise positioning and timing.
The study explores the integration of 5G control-plane capabilities with the BeiDou Navigation Satellite System (BDS) to create a networked spatiotemporal service. This enhances positioning, navigation, and timing capabilities for various applications, including autonomous driving, smart highways, and precision agriculture.
Researchers found that combining Global Navigation Satellite System (GNSS) observations and modern bias products can deliver near-instantaneous centimetre-level positioning. Machine-learning attribution frameworks identified key factors driving remaining errors, providing a data-driven complement to conventional GNSS diagnostics.
Researchers develop two methods to analyze GNSS coordinate time series with missing values, preserving valid information and reducing bias. The methods use linear transformation matrices or marginal likelihood functions to keep cross-gap differences and stabilize non-stationary noise.
Researchers developed a decoupled initialization framework that tackles key challenges in visual-inertial odometry, improving initialization success, accuracy, and efficiency. The method uses aggregated visual constraints, estimates gyroscope bias, and refines states through joint refinement and landmark-free triangulation.
A study comparing archived real-time and post-processed precise satellite products found that real-time streams achieved moderate centimeter-level positioning when corrections were continuously available. However, the gap to post-processed products widened under stricter accuracy targets, especially for single-constellation solutions.
A new study introduces a factor graph optimization framework for integrated GNSS/LEO precise point positioning, combining heterogeneous graph-based stochastic modeling with adaptive covariance rescaling. The method achieved improved three-dimensional positioning accuracy, availability, and short-term continuity in complex environments.
A residual-based machine learning validator enhances GNSS positioning in urban canyons and dense vegetation by reducing wrong fixes and improving correct ambiguity fixing rates. The compact MLP classifier achieves accuracy and precision above 90% in real-time use.
A new study reveals that adding lunar satellites to China's BeiDou-3 navigation constellation can eliminate navigation constellations' drift. The team achieved stable orientation for 60 days and low user range error, demonstrating the effectiveness of lunar satellites as an anchor for the network.
Researchers developed a neural network-based framework that significantly improves the prediction of satellite clock bias for low earth orbit satellites. The approach uses advanced data-processing strategies and a reconstruction fine-tuning mechanism to enable more accurate and stable real-time predictions of satellite timing errors.
Researchers developed a hybrid positioning method combining satellite-broadcast corrections with ground-based network data to maintain centimeter-level accuracy for over 20 minutes after ground corrections are interrupted. The method achieved horizontal and vertical accuracies of 1.30 cm and 5.67 cm in tests using a 300-kilometer network.
A new framework generalizes Gaussian success-rate bounds to a broad family of heavy-tailed distributions, preserving computational simplicity. This enables accurate characterization of robustness against realistic non-Gaussian disturbances in GNSS positioning, urban canyons, autonomous vehicles, and safety-critical applications.
Researchers developed a novel adaptive model, BLAC-Q4DIM, to address ionospheric modeling errors. The model achieved dramatic improvements, reducing errors by 80-85% compared to standard GIMs and 58% compared to fixed-parameter Q4DIM, making it suitable for high-precision navigation and space weather monitoring.
Researchers developed a learning-based navigation system that adapts to changing environments and optimizes sensor fusion. The system uses a quantitative belief state to evaluate the reliability of each sensor, feeding this information into a decision agent that dynamically adjusts fusion strategies.
Researchers use LEO satellites to improve BeiDou-3 Navigation Satellite System (BDS-3) orbit and clock products, cutting positioning time by up to 79.2% and enhancing tracking coverage to over 99%. The approach also accelerates precise point positioning with five LEO satellites.
A new sparse coding method reduces storage and transmission costs for high-precision Global Navigation Satellite System (GNSS) atmospheric corrections. The approach preserves accuracy while minimizing computational resources, making it suitable for global deployment in remote areas.
A new compact ionospheric model reduces unknown parameters while retaining ionospheric correction, strengthening ambiguity estimation and supporting earlier reliable fixing. Tests show faster initialization, higher ambiguity-fixing success rates, and improved positioning accuracy over medium-to-long baselines.
A new framework, FAIRS, improves responsive and statistical bounding performance of uncertainty maps by accounting for non-Gaussian residual behavior. It enhances the responsiveness and reliability of ionospheric corrections in high-precision positioning applications.
Researchers developed an attention-guided Transformer model that learns unique strengths of each satellite mission, preserving differences in constellation design and observation geometry. The approach achieved high correlation coefficients and low RMSE values against references, especially in arid regions.
Researchers developed a new monitoring method for Sea-Based Joint Precision Approach and Landing System (SB-JPALS) using multiple reference receivers and a Boolean collaborative decision rule to detect ionospheric gradients. The approach improved monitoring sensitivity by 12.0% compared to conventional methods.
Researchers have developed a systematic cross-calibration method for GPS satellite data, producing a long-term dataset that spans two full solar cycles. The calibrated dataset provides a reliable resource for studying relativistic electrons in medium Earth orbit, enabling more accurate predictions of electron flux enhancements.
A new approach embeds fundamental physical laws into neural networks, enabling detection of GNSS spoofing with high accuracy. The framework outperforms standard models on unseen attack scenarios, offering a blueprint for building AI-based security systems that anchor decisions in physical invariants.
Researchers have developed a fully automated framework that detects GPS flex power boosts with near-perfect accuracy and maps their geographic centers. The framework, combining detection and autonomous center-fitting systems, replaces slow manual methods and enables real-time monitoring and automated spatial mapping.
A new positioning framework combines satellite navigation with asynchronous ground-based transmitters, enabling faster convergence and improved accuracy. The findings suggest that existing terrestrial communication infrastructure can be repurposed to enhance next-generation navigation services.
Researchers developed a novel navigation method that reads a building's unique magnetic fingerprint to determine its location with unprecedented accuracy. The system uses an array of magnetometers and inertial measurement units to track movement without any external signals, achieving a horizontal positioning RMSE below 1.27 meters.
Researchers developed a layered sound speed gradient model to capture horizontal variations in the upper and deeper layers of the ocean. The new method significantly reduces positioning errors, achieving millimeter-level accuracy in simulations and field experiments.
Researchers developed a cathode-modulated vacuum/air-channel electron tube that eliminates gate leakage current, enabling vacuum tubes to function in integrated circuits. The device operates at room temperature and atmospheric pressure, offering non-saturating output characteristics and potential for high-frequency applications.
Researchers at UCF–UF develop a gram-scale mechanical resonator suspended entirely by diamagnetic levitation, eliminating mechanical supports and energy losses. The device achieves exceptional stability and low dissipation, outperforming high-end MEMS sensors.
Researchers developed a chemo-mechanical switch that consumes virtually no power until detecting gas leaks, offering a clean alternative to traditional sensors. The device can detect hydrogen concentrations as low as 0.3% and achieves an on/off current ratio exceeding 135,000.
Researchers discovered that long-term exposure to high viscosity environments makes glioblastoma cells smaller and more deformable, allowing them to invade healthy tissue more effectively. The study found that viscosity is not just a passive barrier but an active driver of aggressive behavior.
A new method combines holographic optical tweezers with AC electric fields to pre-align and trap nanowires with improved stability and efficiency. This hybrid approach enables predictable, programmable movement, turning random motion into controlled assembly tasks.
Researchers investigate how phase errors in MEMS gyroscopes affect force-to-rebalance rate measurement, identifying which errors matter most and proposing calibration strategies. The study shows that some errors mainly disturb drive-frequency locking, while others degrade scale factor and zero-rate output in the sense loop.
Researchers introduce a new quality-control framework that detects and corrects errors in satellite orbit, clock, and bias information, improving the reliability and availability of real-time high-precision positioning. The method shows improved results in ambiguity fixing rate, incorrect fixing rate, and convergence rate.
A new camera-only visual odometry system uses prebuilt colored point cloud maps to deliver accurate and robust localization in GNSS-challenged environments. The system reduced absolute trajectory error by up to 95% compared to existing methods, while maintaining near real-time efficiency.
Researchers have developed a complementary perception strategy combining visual localization with tactile mapping for robots to locate and read object surface features. The system uses a single RGB-Depth camera and soft pressure sensor array to identify object position, size, and geometry through vision and tactile scanning, respectively.
Researchers developed a Variational Level Set Autoencoder (VLSet-AE) to automate contour recognition in SEM cross-sections of DRIE structures. The model achieved high precision, recognizing critical structural features with low average prediction error of 3.65% and correlation coefficient of 0.998.
Researchers have developed a new method to detect single micro/nanoparticles and protein binding events using non-linear resonance in a commercial quartz crystal microbalance. The system achieves femtogram-level sensitivity without requiring surface functionalization or nanomaterial integration.
A wireless ureteral stent sleeve, UroSleeve, tracks intrarenal pressure through resonance shifts, detecting pressure changes linked to hydronephrosis in an ex vivo porcine kidney model. The system could enable continuous remote follow-up and reduce dependence on episodic radiographic imaging.
A new multi-layer ionospheric mapping function for PPP-AR has been validated, improving convergence times by 4-10% and early-stage positioning. The method outperforms conventional Single Layer Model (SLM) in equatorial regions where ionospheric gradients are hardest to tame.
Researchers have developed a lead-free thin film that significantly improves the efficiency of microdevices in harvesting energy from ambient motion. The Mn-doped bismuth ferrite film exhibits stronger piezoelectric behavior, lower dielectric loss, and improved device-level performance.
Researchers have developed a new artificial intelligence framework called CLAK that enables drones to localize themselves in GPS-denied environments using non-visual sensors such as LiDAR, barometric altitude, and inertial measurements. The model improves localization accuracy while remaining lightweight enough for practical deployment.
Researchers developed a double-slit plasmonic platform-based fiber probe that combines easier light excitation, stronger tip enhancement, broadband stability, and controllable fabrication. The probe achieved 28.6 nm optical imaging resolution under ambient conditions and resolved structures smaller than the diffraction limit.
Researchers from Purdue University and Menlo Microsystems developed a commercial microelectromechanical switch that operates reliably at cryogenic temperatures. The device showed lower operating voltage, lower on-resistance, and strong radio-frequency performance.
A new soft sensing system allows humanoid robots to perceive complex finger posture in real time, enabling precise movement and dexterous manipulation. The system features an omnidirectional bending sensor that tracks pitch and yaw at the finger joints, providing stability and repeatability for demanding actions.
A compact chip holder with integrated electrodes enables temperature control, electric actuation, pressure handling, and optical readout for nanofluidic systems. The platform supports studies on molecular transport, catalytic reactions, and protein aggregation in confined environments.
Researchers developed microfluidic synthesis of polymer microspheres with tunable shape and surface area, enhancing metal loading, mass transfer, and synergistic catalysis. Open-hole Ag-Pt microspheres delivered the strongest performance in converting toxic pollutants into valuable products.
Regional ionospheric structures in the Asian sector can strongly alter satellite navigation positioning, with steep TEC gradients and storm-driven post-sunset irregularities emerging as major sources of error. This study highlights the need for improved global ionosphere models to accurately predict positioning outcomes.
Researchers have developed a compact chemiresistive biosensor that directly transduces biochemical reactions into electrical signals, detecting creatinine concentrations from 1 to 300 mg/dL with high sensitivity and selectivity. The sensor's two-electrode design eliminates the need for reference electrodes and operates in just 35 seconds.
A new navigation integrity method has been developed to handle non-Gaussian observation errors in ocean positioning. The approach assigns differentiated bounds across elevation intervals, allowing protection levels to better track real positioning errors.
Researchers developed a unified detection framework to identify flex power operations in GPS and BeiDou, analyzing how these signal changes affect positioning accuracy and navigation reliability. The study introduces resilient estimation strategies and improved data-processing models to maintain stable navigation services.