Yueyang Ben
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Yueyang Ben has a total of 61 co-authors. Here are the top co-authors by number of shared publications:
- Qian Li
College of Intelligent Systems Science and Engineering, Harbin Engineering University, Harbin, China
17 shared publications - Xinle Zang
College of Intelligent Science and Engineering, Harbin Engineering University, Harbin, China
4 shared publications - Yuxin Zhao
College of Intelligent Systems Science and Engineering, Harbin Engineering University, Harbin, China
4 shared publications - Chen Lv
College of Intelligent Systems Science and Engineering, Harbin Engineering University, Harbin, China
3 shared publications - Hongdian Huang
College of Intelligent Systems Science and Engineering, Harbin Engineering University, Harbin, China
3 shared publications - Jiancheng Wang
College of Intelligent Science and Engineering, Harbin Engineering University, Harbin, China
3 shared publications
Latest Publications
- 26 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 75  (00189456) , 8508816
Scenario-Aided Robust Factor Graph Optimization for GNSS/INS Integration in Complex Environments
The integration of global navigation satellite systems (GNSS) and inertial navigation systems (INS) employs robust methods to ensure reliable navigation solutions in challenging environments, where the robustness parameter is used to balance accuracy and estimator stability. However, in time-varying GNSS conditions or prolonged harsh environments, a parameter–environment inconsistency may lead to incorrect rejection or excessive down-weighting of normal measurements, causing accuracy degradation or divergence. To address this issue, this paper proposes a scenario-aided robust factor graph optimization (RFGO) algorithm that adapts the robustness parameter to time-varying conditions. Scenarios are first classified based on GNSS/INS integrated error characteristics, and an observation model is constructed using GNSS signal features. Considering the latency of the integration scenarios induced by INS characteristics, a hidden semi-Markov model (HSMM) with state duration modeling is employed for scenario awareness. Based on the inferred scenarios, a scenario-driven robustness adjustment strategy is developed within the dynamic covariance scaling-based RFGO framework, thereby ensuring consistency between the robustness parameter and the actual scenarios. Vehicle experiment results illustrated that the HSMM achieved recognition accuracies of 80.24%, 75.48%, and 96.55% in three different scenarios, respectively. Compared with fixed-parameter methods, the proposed method increased the proportion of high-accuracy epochs by 13.6%–27.1% and reduced abnormal epochs by 41.4%–51.4%, demonstrating improved accuracy and solution stability in complex environments.
- 26 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 75  (00189456) , 8508816
Scenario-Aided Robust Factor Graph Optimization for GNSS/INS Integration in Complex Environments
The integration of global navigation satellite systems (GNSS) and inertial navigation systems (INS) employs robust methods to ensure reliable navigation solutions in challenging environments, where the robustness parameter is used to balance accuracy and estimator stability. However, in time-varying GNSS conditions or prolonged harsh environments, a parameter–environment inconsistency may lead to incorrect rejection or excessive down-weighting of normal measurements, causing accuracy degradation or divergence. To address this issue, this paper proposes a scenario-aided robust factor graph optimization (RFGO) algorithm that adapts the robustness parameter to time-varying conditions. Scenarios are first classified based on GNSS/INS integrated error characteristics, and an observation model is constructed using GNSS signal features. Considering the latency of the integration scenarios induced by INS characteristics, a hidden semi-Markov model (HSMM) with state duration modeling is employed for scenario awareness. Based on the inferred scenarios, a scenario-driven robustness adjustment strategy is developed within the dynamic covariance scaling-based RFGO framework, thereby ensuring consistency between the robustness parameter and the actual scenarios. Vehicle experiment results illustrated that the HSMM achieved recognition accuracies of 80.24%, 75.48%, and 96.55% in three different scenarios, respectively. Compared with fixed-parameter methods, the proposed method increased the proportion of high-accuracy epochs by 13.6%–27.1% and reduced abnormal epochs by 41.4%–51.4%, demonstrating improved accuracy and solution stability in complex environments.
- 12 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 26  (4) , 5981-5994
A Pseudo-GNSS Reconstruction Method of INS Based on the AT-DCLSTM
Global Navigation Satellite System (GNSS) and inertial navigation system (INS) together overcome their individual shortcomings, providing accurate navigation results in GNSS/INS integrated systems. However, the entire system’s positioning accuracy deteriorates over time during GNSS outages. In this article, a pseudo-GNSS measurement is predicted by the combination of an algorithm based on a dual-channel long short-term memory (DCLSTM) neural network and an attention mechanism (AT) model for aiding INS. Considering the differences in the propagation characteristics of longitude and latitude errors, the normal LSTM neural network is optimized channel-wise to predict longitude and latitude errors, respectively, and get the pseudo-GNSS as the observation. In addition, the AT is used to improve the ability of the LSTM network to capture the characteristics of long-term sequence information. Then, the pseudo-GNSS measurement update model for Kalman filter (KF) is established to incorporate these predictions into GNSS/INS integration during GNSS outages. Finally, experiments based on sea trials are implemented to evaluate the proposed method. Trial results show that the proposed method can effectively improve the performance of the integrated navigation system during GNSS outages.
- 12 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 26  (4) , 5981-5994
A Pseudo-GNSS Reconstruction Method of INS Based on the AT-DCLSTM
Global Navigation Satellite System (GNSS) and inertial navigation system (INS) together overcome their individual shortcomings, providing accurate navigation results in GNSS/INS integrated systems. However, the entire system’s positioning accuracy deteriorates over time during GNSS outages. In this article, a pseudo-GNSS measurement is predicted by the combination of an algorithm based on a dual-channel long short-term memory (DCLSTM) neural network and an attention mechanism (AT) model for aiding INS. Considering the differences in the propagation characteristics of longitude and latitude errors, the normal LSTM neural network is optimized channel-wise to predict longitude and latitude errors, respectively, and get the pseudo-GNSS as the observation. In addition, the AT is used to improve the ability of the LSTM network to capture the characteristics of long-term sequence information. Then, the pseudo-GNSS measurement update model for Kalman filter (KF) is established to incorporate these predictions into GNSS/INS integration during GNSS outages. Finally, experiments based on sea trials are implemented to evaluate the proposed method. Trial results show that the proposed method can effectively improve the performance of the integrated navigation system during GNSS outages.
- 10 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 26  (2) , 1819-1831
A Novel SINS/EM Log Integrated Navigation Method Based on Motion Constraints
The Strapdown Inertial Navigation System (SINS)/electromagnetic log (EM Log) integrated navigation system is widely employed on surface vessels and submarines. However, the EM Log measures only the fore-and-aft velocity of the vehicle relative to seawater, which is different from the vehicle’s true velocity direction due to the influence of pitch, roll, and sideslip, thereby degrading navigation accuracy. To address this issue, this paper proposes a motion-constrained SINS/EM Log integrated navigation method. This method uses the motion characteristics of marine vehicles to establish appropriate constraint conditions. Subsequently, constraint equations are formulated to reconstruct the sway and heave velocities that cannot be measured by the EM Log, as well as using the acceleration and heading rate to identify the turning maneuver and then correct the sideslip. The installation misalignment angles between SINS and EM Log are augmented to the state vector of the Kalman filter with regard to their negative effects on constraint conditions, and the error equations with SINS misalignment angles and EM Log installation misalignment angles under motion constraints are derived. The final results of simulation and experiment demonstrate that, compared with the augmented dimension method, the proposed method effectively suppresses the velocity error during the turning maneuvers and improves the overall system performance.
- 10 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 26  (2) , 1819-1831
A Novel SINS/EM Log Integrated Navigation Method Based on Motion Constraints
The Strapdown Inertial Navigation System (SINS)/electromagnetic log (EM Log) integrated navigation system is widely employed on surface vessels and submarines. However, the EM Log measures only the fore-and-aft velocity of the vehicle relative to seawater, which is different from the vehicle’s true velocity direction due to the influence of pitch, roll, and sideslip, thereby degrading navigation accuracy. To address this issue, this paper proposes a motion-constrained SINS/EM Log integrated navigation method. This method uses the motion characteristics of marine vehicles to establish appropriate constraint conditions. Subsequently, constraint equations are formulated to reconstruct the sway and heave velocities that cannot be measured by the EM Log, as well as using the acceleration and heading rate to identify the turning maneuver and then correct the sideslip. The installation misalignment angles between SINS and EM Log are augmented to the state vector of the Kalman filter with regard to their negative effects on constraint conditions, and the error equations with SINS misalignment angles and EM Log installation misalignment angles under motion constraints are derived. The final results of simulation and experiment demonstrate that, compared with the augmented dimension method, the proposed method effectively suppresses the velocity error during the turning maneuvers and improves the overall system performance.
- 24 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 26  (1) , 791-803
Physically Guided and Perceptually Adaptive Fusion for Underwater Image Enhancement in Visual SLAM
Underwater visual Simultaneous Localization and Mapping (SLAM) suffers from severe image degradation caused by wavelength-dependent absorption, scattering, and turbidity, which significantly impair feature detection and tracking. While existing enhancement methods can improve perceptual image quality, they often fail to preserve structural details essential for robust SLAM performance. To bridge this gap, we propose a unified enhancement framework that jointly considers physical imaging models, perceptual consistency, and SLAM-specific feature requirements. The framework incorporates a structure-aware red channel prior for more accurate transmission estimation, a feature-adaptive Retinex-based strategy for local brightness correction, and a confidence-guided color compensation mechanism that maintains chromatic reliability across regions. To ensure downstream compatibility, we further design a SLAM-aware multi-feature fusion module that adaptively integrates perceptual and structural cues to enhance both visual fidelity and feature stability. Extensive experiments on the SUIM dataset and real underwater SLAM sequences demonstrate that our method achieves the highest average UIQM (0.6880) and UCIQE (34.9839), outperforming six representative enhancement techniques. The enhanced images also boost SLAM accuracy and robustness, increasing feature matches and reducing ATE RMSE and loop drift, while parameter analysis confirms stable performance across varying underwater conditions. These results confirm the framework’s effectiveness in enhancing underwater perception and its practical utility for SLAM-based navigation and mapping.
- 24 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 26  (1) , 791-803
Physically Guided and Perceptually Adaptive Fusion for Underwater Image Enhancement in Visual SLAM
Underwater visual Simultaneous Localization and Mapping (SLAM) suffers from severe image degradation caused by wavelength-dependent absorption, scattering, and turbidity, which significantly impair feature detection and tracking. While existing enhancement methods can improve perceptual image quality, they often fail to preserve structural details essential for robust SLAM performance. To bridge this gap, we propose a unified enhancement framework that jointly considers physical imaging models, perceptual consistency, and SLAM-specific feature requirements. The framework incorporates a structure-aware red channel prior for more accurate transmission estimation, a feature-adaptive Retinex-based strategy for local brightness correction, and a confidence-guided color compensation mechanism that maintains chromatic reliability across regions. To ensure downstream compatibility, we further design a SLAM-aware multi-feature fusion module that adaptively integrates perceptual and structural cues to enhance both visual fidelity and feature stability. Extensive experiments on the SUIM dataset and real underwater SLAM sequences demonstrate that our method achieves the highest average UIQM (0.6880) and UCIQE (34.9839), outperforming six representative enhancement techniques. The enhanced images also boost SLAM accuracy and robustness, increasing feature matches and reducing ATE RMSE and loop drift, while parameter analysis confirms stable performance across varying underwater conditions. These results confirm the framework’s effectiveness in enhancing underwater perception and its practical utility for SLAM-based navigation and mapping.
- 10 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25  (16) , 31086-31096
Robust Initialization of Underwater Monocular Visual–Inertial–Pressure Odometry
In the underwater environment, image degradation and insufficient IMU excitation make the initialization of visual-inertial odometry a serious challenge. The traditional method improves the performance of initialization by fusing depth information, but does not consider the camera-depth misalignment effect, and the estimation accuracy of gyroscope bias is affected by inaccurate visual Structure-from-Motion (SfM). To address these issues, this article proposes a robust initialization method by introducing the robust cost function and depth information to provide a high-quality initial state for Underwater Monocular Visual-Inertial-Pressure Odometry (UM-VIPO). Our method begins with a vision-only structure. The Truncated Least Squares (TLS) algorithm is used to suppress outliers in feature matching, thereby significantly improving the accuracy and robustness of visual SfM. In addition, the Graduated Non-Convexity (GNC) algorithm is used to solve the non-convex optimization problem induced by the TLS, effectively avoiding convergence to local minima. During visual-depth alignment, depth measurements are introduced to recover the metric scale of the visual poses. By aligning the visual poses and depth measurements with the world coordinate system, the camera-depth misalignment effect is effectively mitigated. In the visual-inertial alignment phase, the gyroscope bias is estimated from the vision-only structure, while the IMU pre-integration is leveraged to calculate the initial velocity in the world coordinate system. Then, all initialized parameters are used to execute a local Bundle Adjustment (BA). Finally, the proposed method is validated on the AQUALOC dataset, demonstrating its effectiveness and superiority.
- 10 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25  (16) , 31086-31096
Robust Initialization of Underwater Monocular Visual–Inertial–Pressure Odometry
In the underwater environment, image degradation and insufficient IMU excitation make the initialization of visual-inertial odometry a serious challenge. The traditional method improves the performance of initialization by fusing depth information, but does not consider the camera-depth misalignment effect, and the estimation accuracy of gyroscope bias is affected by inaccurate visual Structure-from-Motion (SfM). To address these issues, this article proposes a robust initialization method by introducing the robust cost function and depth information to provide a high-quality initial state for Underwater Monocular Visual-Inertial-Pressure Odometry (UM-VIPO). Our method begins with a vision-only structure. The Truncated Least Squares (TLS) algorithm is used to suppress outliers in feature matching, thereby significantly improving the accuracy and robustness of visual SfM. In addition, the Graduated Non-Convexity (GNC) algorithm is used to solve the non-convex optimization problem induced by the TLS, effectively avoiding convergence to local minima. During visual-depth alignment, depth measurements are introduced to recover the metric scale of the visual poses. By aligning the visual poses and depth measurements with the world coordinate system, the camera-depth misalignment effect is effectively mitigated. In the visual-inertial alignment phase, the gyroscope bias is estimated from the vision-only structure, while the IMU pre-integration is leveraged to calculate the initial velocity in the world coordinate system. Then, all initialized parameters are used to execute a local Bundle Adjustment (BA). Finally, the proposed method is validated on the AQUALOC dataset, demonstrating its effectiveness and superiority.
- 2 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25  (15) , 29296-29309
Robust Factor Graph Optimization for GNSS/INS Tightly Coupled Integration: A Flexible Strategy for Urban Navigation Resilience
Intelligent transportation systems, including autonomous vehicles, require precise and reliable navigation and positioning systems to ensure safety. However, global navigation satellite system (GNSS) observations are degraded by signal interference, such as obstructions and reflections from high-rise buildings. Navigation methods struggle to effectively process GNSS outliers caused by multipath effects and non-line-of-sight (NLOS) reception in urban environments under changing scenarios. To address this issue, this study proposes a robust factor graph optimization (FGO) approach based on a flexible strategy. Initially, a quality evaluation method utilizing the Mahalanobis distance of pseudorange residual test statistics is developed, enabling smooth and dynamic adjustment of the measurement fusion weights for each satellite. Furthermore, by integrating both satellite quantity and quality data, a flexible strategy is proposed that fully accounts for the differences in measurement quality and the reference values of each satellite across different scenarios. This integration with the navigation environment improves the robustness of the tightly coupled navigation system. Three vehicle-mounted experiments in urban canyon environments were conducted, and the results indicate that, compared to conventional methods, the root mean square errors of horizontal positioning were reduced by 13.24%, 26.06%, and 33.92%, respectively. The proposed method effectively mitigates the impact of GNSS outliers and enhances the reliability of GNSS-based navigation systems for applications such as autonomous vehicles, unmanned aerial vehicles (UAVs) and urban transportation systems.
- 2 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25  (15) , 29296-29309
Robust Factor Graph Optimization for GNSS/INS Tightly Coupled Integration: A Flexible Strategy for Urban Navigation Resilience
Intelligent transportation systems, including autonomous vehicles, require precise and reliable navigation and positioning systems to ensure safety. However, global navigation satellite system (GNSS) observations are degraded by signal interference, such as obstructions and reflections from high-rise buildings. Navigation methods struggle to effectively process GNSS outliers caused by multipath effects and non-line-of-sight (NLOS) reception in urban environments under changing scenarios. To address this issue, this study proposes a robust factor graph optimization (FGO) approach based on a flexible strategy. Initially, a quality evaluation method utilizing the Mahalanobis distance of pseudorange residual test statistics is developed, enabling smooth and dynamic adjustment of the measurement fusion weights for each satellite. Furthermore, by integrating both satellite quantity and quality data, a flexible strategy is proposed that fully accounts for the differences in measurement quality and the reference values of each satellite across different scenarios. This integration with the navigation environment improves the robustness of the tightly coupled navigation system. Three vehicle-mounted experiments in urban canyon environments were conducted, and the results indicate that, compared to conventional methods, the root mean square errors of horizontal positioning were reduced by 13.24%, 26.06%, and 33.92%, respectively. The proposed method effectively mitigates the impact of GNSS outliers and enhances the reliability of GNSS-based navigation systems for applications such as autonomous vehicles, unmanned aerial vehicles (UAVs) and urban transportation systems.
- 2 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74  (00189456) , 1-10
A Factor Graph With Mercator Projection-Based Cooperative Localization Algorithm for Multiple AUVs
To address the discrepancy between geodetic coordinates provided by navigation equipment on Autonomous Underwater Vehicles (AUVs) and Cartesian coordinates employed for calculating Euclidean distance within a Cooperative Localization (CL) system, a Factor Graph (FG) with Mercator Projection-based CL algorithm for multiple AUVs (MAUVs) is proposed. First, the Mercator projection is utilized to formulate the range measurement equation, and with the kinematic equation of the AUV, a state-space model of a CL system for MAUVs is constructed. Furthermore, the homologous FG model of the CL system is established. Leveraging the independence of the coordinates inherent in the Mercator projection, the FG model avoids the introduction of additional cycles. The Sum-Product Algorithm (SPA) is utilized to calculate navigation states of the slave AUV. Then, a Performance Evaluation Function (PEF) for a CL system with two master AUVs in the Mercator projection-based coordinate system is established, and the consistency of PEFs across different coordinate systems is demonstrated. Finally, both simulation and field test results indicate that the proposed CL algorithm offers superior localization accuracy and lower execution time compared to conventional Extended Kalman Filter (EKF)-based CL algorithms.
- 2 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74  (00189456) , 1-10
A Factor Graph With Mercator Projection-Based Cooperative Localization Algorithm for Multiple AUVs
To address the discrepancy between geodetic coordinates provided by navigation equipment on Autonomous Underwater Vehicles (AUVs) and Cartesian coordinates employed for calculating Euclidean distance within a Cooperative Localization (CL) system, a Factor Graph (FG) with Mercator Projection-based CL algorithm for multiple AUVs (MAUVs) is proposed. First, the Mercator projection is utilized to formulate the range measurement equation, and with the kinematic equation of the AUV, a state-space model of a CL system for MAUVs is constructed. Furthermore, the homologous FG model of the CL system is established. Leveraging the independence of the coordinates inherent in the Mercator projection, the FG model avoids the introduction of additional cycles. The Sum-Product Algorithm (SPA) is utilized to calculate navigation states of the slave AUV. Then, a Performance Evaluation Function (PEF) for a CL system with two master AUVs in the Mercator projection-based coordinate system is established, and the consistency of PEFs across different coordinate systems is demonstrated. Finally, both simulation and field test results indicate that the proposed CL algorithm offers superior localization accuracy and lower execution time compared to conventional Extended Kalman Filter (EKF)-based CL algorithms.
- 28 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74  (00189456) , 1-13
A Novel Calibration Method for SINS and DVL Based on Improved Dung Beetle Optimizer
The calibration quality of the Doppler velocity log (DVL) significantly impacts the performance of the strap-down inertial navigation system (SINS)/ DVL integrated navigation system. Optimizer-based calibration algorithms consider the installation misalignment angle and scale factor error as optimization objectives, thus simplifying the calibration process into an optimizer problem. However, optimizer-based calibration under large installation misalignment angle suffers from slow convergence speed and local optima issues problems. To address these challenges, this paper proposes a novel calibration algorithm for estimating installation misalignment angle and scale factor error between SINS and DVL sensors based on an improved dung beetle optimizer. A novel multi-strategy integration is designed in the dung beetle optimizer algorithm to balance the exploration and exploitation during optimization. Due to fast and accurate optimization, this calibration algorithm streamlines calibration times and removes the necessity for specialized vehicle operations. Both simulation and sea trials demonstrate the capability of the algorithm to rapidly and accurately estimate scale factor error and misalignment angle, thereby significantly improving navigation accuracy.
- 28 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74  (00189456) , 1-13
A Novel Calibration Method for SINS and DVL Based on Improved Dung Beetle Optimizer
The calibration quality of the Doppler velocity log (DVL) significantly impacts the performance of the strap-down inertial navigation system (SINS)/ DVL integrated navigation system. Optimizer-based calibration algorithms consider the installation misalignment angle and scale factor error as optimization objectives, thus simplifying the calibration process into an optimizer problem. However, optimizer-based calibration under large installation misalignment angle suffers from slow convergence speed and local optima issues problems. To address these challenges, this paper proposes a novel calibration algorithm for estimating installation misalignment angle and scale factor error between SINS and DVL sensors based on an improved dung beetle optimizer. A novel multi-strategy integration is designed in the dung beetle optimizer algorithm to balance the exploration and exploitation during optimization. Due to fast and accurate optimization, this calibration algorithm streamlines calibration times and removes the necessity for specialized vehicle operations. Both simulation and sea trials demonstrate the capability of the algorithm to rapidly and accurately estimate scale factor error and misalignment angle, thereby significantly improving navigation accuracy.
- 26 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74  (00189456) , 1-8
A Novel Initial Position Matching Method Utilizing Multibeam Bathymetry for Underwater Vehicles
When using Terrain Aided Navigation (TAN), terrain matching related algorithms typically require high-precision initial position. However, in underwater navigation, Strapdown Inertial Navigation System (SINS) often cannot provide the required accurate initial position information. To address this, this paper proposes a novel initial position matching method utilizing multibeam bathymetry for underwater vehicles, leveraging yaw and depth information to overcome the limitations of initial position error. Firstly, the depth information pre-measured by the Multibeam Echosounder (MBES) is preprocessed. Subsequently, predictive interpolation based on sparse pseudo-input Gaussian processes is performed, and High-Precision Contour Lines (HPCL) are generated using the Marching Squares algorithm. Real-time measurements from three beams via MBES are then individually matched with the HPCL, and the matching results are obtained based on trajectory parallelism. These results are used as the initial point for subsequent matching, and the iterative matching process continues. Ultimately, the final matching result must satisfy the geometric relationship perpendicular to the trajectory to achieve high-precision initial positioning. Furthermore, results from several sets of sea experiments demonstrate the feasibility of proposed algorithm, with matching initial position error consistently within 100m.
- 26 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74  (00189456) , 1-8
A Novel Initial Position Matching Method Utilizing Multibeam Bathymetry for Underwater Vehicles
When using Terrain Aided Navigation (TAN), terrain matching related algorithms typically require high-precision initial position. However, in underwater navigation, Strapdown Inertial Navigation System (SINS) often cannot provide the required accurate initial position information. To address this, this paper proposes a novel initial position matching method utilizing multibeam bathymetry for underwater vehicles, leveraging yaw and depth information to overcome the limitations of initial position error. Firstly, the depth information pre-measured by the Multibeam Echosounder (MBES) is preprocessed. Subsequently, predictive interpolation based on sparse pseudo-input Gaussian processes is performed, and High-Precision Contour Lines (HPCL) are generated using the Marching Squares algorithm. Real-time measurements from three beams via MBES are then individually matched with the HPCL, and the matching results are obtained based on trajectory parallelism. These results are used as the initial point for subsequent matching, and the iterative matching process continues. Ultimately, the final matching result must satisfy the geometric relationship perpendicular to the trajectory to achieve high-precision initial positioning. Furthermore, results from several sets of sea experiments demonstrate the feasibility of proposed algorithm, with matching initial position error consistently within 100m.
- 13 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25  (4) , 7053-7067
Residual Clustering Analysis-Based Fault Detection and Exclusion for GNSS/INS Tightly Coupled Integration
In tightly coupled integrated navigation involving the Inertial Navigation System (INS) and the Global Navigation Satellite System (GNSS), the fault detection and exclusion (FDE) method can effectively ensure navigation accuracy by detecting faults in the residual domain. However, in challenging scenarios, the FDE method will degrade rapidly due to error accumulation, measurement faults, and receiver clock bias instability. To address this issue, a novel FDE method based on residual clustering analysis is proposed. By constructing the Euclidean distance matrix of measurement residuals, the cluster density of measurement residuals is calculated, eliminating the common bias caused by inaccurate estimation of receiver clock or unmodeled states. Furthermore, to enhance the robustness of the proposed method, the fault detection threshold is dynamically modified according to cluster density and position error covariance, and the degradation of the FDE caused by position error is suppressed. Experimental demonstrations confirm the effectiveness and advantages of the proposed method. Compared with conventional FDE methods, simulation results show that the mean error of horizontal position decreased by 82.25% to 83.49% in challenging environments. In field tests, the mean error of horizontal position was reduced by 29.61% to 47.87% and 17.66% to 64.94%, respectively.
- 13 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25  (4) , 7053-7067
Residual Clustering Analysis-Based Fault Detection and Exclusion for GNSS/INS Tightly Coupled Integration
In tightly coupled integrated navigation involving the Inertial Navigation System (INS) and the Global Navigation Satellite System (GNSS), the fault detection and exclusion (FDE) method can effectively ensure navigation accuracy by detecting faults in the residual domain. However, in challenging scenarios, the FDE method will degrade rapidly due to error accumulation, measurement faults, and receiver clock bias instability. To address this issue, a novel FDE method based on residual clustering analysis is proposed. By constructing the Euclidean distance matrix of measurement residuals, the cluster density of measurement residuals is calculated, eliminating the common bias caused by inaccurate estimation of receiver clock or unmodeled states. Furthermore, to enhance the robustness of the proposed method, the fault detection threshold is dynamically modified according to cluster density and position error covariance, and the degradation of the FDE caused by position error is suppressed. Experimental demonstrations confirm the effectiveness and advantages of the proposed method. Compared with conventional FDE methods, simulation results show that the mean error of horizontal position decreased by 82.25% to 83.49% in challenging environments. In field tests, the mean error of horizontal position was reduced by 29.61% to 47.87% and 17.66% to 64.94%, respectively.