Xin Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Xin Zhang has a total of 43 co-authors. Here are the top co-authors by number of shared publications:
- Bo He
School of Information Science and Engineering, Ocean University of China, Qingdao, China
8 shared publications - Xiaokai Mu
Qingdao Innovation and Development Base, Harbin Engineering University, Qingdao, China
5 shared publications - Ping Qin
College of Information Science and Engineering, Ocean University of China, Qingdao 266100, China
2 shared publications - Shuang Gao
Ocean University of China, Qingdao, CN, Qingdao, China
2 shared publications - Tianhong Yan
School of Mechatronic Engineering and Automation, Shanghai University, Shanghai, China
2 shared publications - Chen Ning
School of Computer and Electronic Information, Nanjing Normal University, Nanjing, China
1 shared publication
Latest Publications
- 13 October 2026
- Published by Wiley in Journal of Field Robotics
Context‐Aware Reconstruction for DVL Outlier Detection in Robust AUV Navigation
In complex shallow‐water environments, environmental and acoustic interferences may lead to anomalous Doppler Velocity Log (DVL) measurements, significantly degrading Autonomous Underwater Vehicle (AUV) navigation performance. To address this issue, this article proposes a robust AUV navigation strategy based on context‐aware reconstruction learning. First, a Context‐Aware Reconstruction Network (CAR‐Net) based on deep Autoencoder (AE) architecture is proposed for DVL outlier detection. By constructing a deep network composed of a heterogeneous AE, a multi‐head self‐attention mechanism, and multi‐scale dilated residual convolutions, the model is capable of achieving global perception over the entire temporal domain while progressively and accurately reconstructing fine‐grained local temporal features. Subsequently, a context‐aware reconstruction mechanism is introduced, which incorporates historical data as contextual constraints, thereby enabling focused modeling of recent time‐series data. Meanwhile, to enhance the ability to fit critical recent moments, a time‐weighted loss is proposed, increasing sensitivity to predictions at the end of the sequence and thereby further improving the reconstruction accuracy of complex temporal patterns. Real‐world AUV sea trial data thoroughly validate that the Extended Kalman Filter (EKF) framework integrated with CAR‐Net demonstrates outstanding performance in extreme outlier detection and robust navigation, providing a novel deep‐learning paradigm for AUV operations in complex underwater environments.
- 13 October 2026
- Published by Wiley in Journal of Field Robotics
Context‐Aware Reconstruction for DVL Outlier Detection in Robust AUV Navigation
In complex shallow‐water environments, environmental and acoustic interferences may lead to anomalous Doppler Velocity Log (DVL) measurements, significantly degrading Autonomous Underwater Vehicle (AUV) navigation performance. To address this issue, this article proposes a robust AUV navigation strategy based on context‐aware reconstruction learning. First, a Context‐Aware Reconstruction Network (CAR‐Net) based on deep Autoencoder (AE) architecture is proposed for DVL outlier detection. By constructing a deep network composed of a heterogeneous AE, a multi‐head self‐attention mechanism, and multi‐scale dilated residual convolutions, the model is capable of achieving global perception over the entire temporal domain while progressively and accurately reconstructing fine‐grained local temporal features. Subsequently, a context‐aware reconstruction mechanism is introduced, which incorporates historical data as contextual constraints, thereby enabling focused modeling of recent time‐series data. Meanwhile, to enhance the ability to fit critical recent moments, a time‐weighted loss is proposed, increasing sensitivity to predictions at the end of the sequence and thereby further improving the reconstruction accuracy of complex temporal patterns. Real‐world AUV sea trial data thoroughly validate that the Extended Kalman Filter (EKF) framework integrated with CAR‐Net demonstrates outstanding performance in extreme outlier detection and robust navigation, providing a novel deep‐learning paradigm for AUV operations in complex underwater environments.
- 19 March 2026
- Published by Elsevier in Ocean Engineering
- Vol. 355
A Full-Mission integrated navigation framework for Shallow-Sea AUV based on GPR-Assisted EKF
- 19 March 2026
- Published by Elsevier in Ocean Engineering
- Vol. 355
A Full-Mission integrated navigation framework for Shallow-Sea AUV based on GPR-Assisted EKF
- Open Access
- 24 May 2023
- Published by MDPI AG in Electronics
- Vol. 12 (11) , 2372
Adaptive SLAM Methodology Based on Simulated Annealing Particle Swarm Optimization for AUV Navigation
Simultaneous localization and mapping (SLAM) is crucial and challenging for autonomous underwater vehicle (AUV) autonomous navigation in complex and uncertain ocean environments. However, inaccurate time-varying observation noise parameters may lead to filtering divergence and poor mapping accuracy. In addition, particles are easily trapped in local extrema during the resampling, which may lead to inaccurate state estimation. In this paper, we propose an innovative simulated annealing particle swarm optimization-adaptive unscented FastSLAM (SAPSO-AUFastSLAM) algorithm. To cope with the unknown observation noise, the maximum a posteriori probability estimation algorithm is introduced into SLAM to recursively correct the measurement noise. Firstly, the Sage–Husa (SH) based unscented particle filter (UPF) algorithm is proposed to estimate time-varying measurement noise adaptively in AUV path estimation for improving filtering accuracy. Secondly, the SH-based unscented Kalman filter (UKF) algorithm is proposed to enhance mapping accuracy in feature estimation. Thirdly, SAPSO-based resampling is proposed to optimize posterior particles. The random judgment mechanism is used to update feasible solutions iteratively, which makes particles disengage local extreme values and achieve optimal global effects. The effectiveness and accuracy of the proposed algorithm are evaluated through simulation and sea trial data. The average AUV navigation accuracy of the presented SAPSO-AUFastSLAM method is improved by 18.0% compared to FastSLAM, 6.5% compared to UFastSLAM, and 5.9% compared to PSO-UFastSLAM.
- 24 May 2023
- Published by MDPI AG in Electronics
- Vol. 12 (11) , 2372
Open AccessAdaptive SLAM Methodology Based on Simulated Annealing Particle Swarm Optimization for AUV Navigation
Simultaneous localization and mapping (SLAM) is crucial and challenging for autonomous underwater vehicle (AUV) autonomous navigation in complex and uncertain ocean environments. However, inaccurate time-varying observation noise parameters may lead to filtering divergence and poor mapping accuracy. In addition, particles are easily trapped in local extrema during the resampling, which may lead to inaccurate state estimation. In this paper, we propose an innovative simulated annealing particle swarm optimization-adaptive unscented FastSLAM (SAPSO-AUFastSLAM) algorithm. To cope with the unknown observation noise, the maximum a posteriori probability estimation algorithm is introduced into SLAM to recursively correct the measurement noise. Firstly, the Sage–Husa (SH) based unscented particle filter (UPF) algorithm is proposed to estimate time-varying measurement noise adaptively in AUV path estimation for improving filtering accuracy. Secondly, the SH-based unscented Kalman filter (UKF) algorithm is proposed to enhance mapping accuracy in feature estimation. Thirdly, SAPSO-based resampling is proposed to optimize posterior particles. The random judgment mechanism is used to update feasible solutions iteratively, which makes particles disengage local extreme values and achieve optimal global effects. The effectiveness and accuracy of the proposed algorithm are evaluated through simulation and sea trial data. The average AUV navigation accuracy of the presented SAPSO-AUFastSLAM method is improved by 18.0% compared to FastSLAM, 6.5% compared to UFastSLAM, and 5.9% compared to PSO-UFastSLAM.
- 24 February 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 63-69
Trajectory Tracking of a Quadrotor Underwater Delivery Platform via Adaptive Sliding Mode Control
A new Quadrotor Underwater Delivery Platform (QUDP) was proposed for underwater transportation, in which its configuration has been well used in air quadrotors. But there are still new challenges to controlling such robots underwater. In this paper, a complete dynamics model of QUDP is established first. Then, the trajectory tracker is designed as the control outer loop based on the adaptive sliding mode (ASMC) under an unknown time-varying load, and the stability is proved by using Lyapunov stability theory. Meanwhile, the PD controller is designed as the control inner loop to track the change of attitude angle. Finally, it is verified by numerical simulation. The results show that the proposed controller can perform well with an unknown load and improve the robustness of the system.
- 24 February 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 63-69
Trajectory Tracking of a Quadrotor Underwater Delivery Platform via Adaptive Sliding Mode Control
A new Quadrotor Underwater Delivery Platform (QUDP) was proposed for underwater transportation, in which its configuration has been well used in air quadrotors. But there are still new challenges to controlling such robots underwater. In this paper, a complete dynamics model of QUDP is established first. Then, the trajectory tracker is designed as the control outer loop based on the adaptive sliding mode (ASMC) under an unknown time-varying load, and the stability is proved by using Lyapunov stability theory. Meanwhile, the PD controller is designed as the control inner loop to track the change of attitude angle. Finally, it is verified by numerical simulation. The results show that the proposed controller can perform well with an unknown load and improve the robustness of the system.
- Open Access
- 21 October 2022
- Published by MDPI AG in Journal of Marine Science and Engineering
- Vol. 10 (10) , 1563
A Variational Bayesian-Based Simultaneous Localization and Mapping Method for Autonomous Underwater Vehicle Navigation
Simultaneous Localization and Mapping (SLAM) is a well-known solution for mapping and realizing autonomous navigation of an Autonomous Underwater Vehicle (AUV) in unknown underwater environments. However, the inaccurate time-varying observation noise will cause filtering divergence and reduce the accuracy of localization and feature estimation. In this paper, VB-AUFastSLAM based on the unscented-FastSLAM (UFastSLAM) and the Variational Bayesian (VB) is proposed. The UFastSLAM combines unscented particle filter (UPF) and unscented Kalman filter (UKF) to estimate the AUV poses and features. In addition, to resist the unknown time-varying observation noise, the method of Variational Bayesian learning is introduced into the SLAM framework. Firstly, the VB method is used to estimate the joint posterior probability of the AUV path and observation noise. The Inverse-Gamma distribution is used to model the observation noise and real-time noise parameters estimation is performed to improve the AUV localization accuracy. Secondly, VB is reused to estimate the noise parameters in the feature update stage to enhance the performance of the feature estimation. The proposed algorithms are first validated in an open-source simulation environment. Then, an AUV SLAM system based on the Inertial Navigation System (INS), Doppler Velocity Log (DVL), and single-beam Sonar are also built to verify the effectiveness of the proposed algorithms in the marine environment. The accuracy of the proposed methods can reach 0.742% and 0.776% of the range, respectively, which is much better than 1.825% and 1.397% of the traditional methods.
- 21 October 2022
- Published by MDPI AG in Journal of Marine Science and Engineering
- Vol. 10 (10) , 1563
Open AccessA Variational Bayesian-Based Simultaneous Localization and Mapping Method for Autonomous Underwater Vehicle Navigation
Simultaneous Localization and Mapping (SLAM) is a well-known solution for mapping and realizing autonomous navigation of an Autonomous Underwater Vehicle (AUV) in unknown underwater environments. However, the inaccurate time-varying observation noise will cause filtering divergence and reduce the accuracy of localization and feature estimation. In this paper, VB-AUFastSLAM based on the unscented-FastSLAM (UFastSLAM) and the Variational Bayesian (VB) is proposed. The UFastSLAM combines unscented particle filter (UPF) and unscented Kalman filter (UKF) to estimate the AUV poses and features. In addition, to resist the unknown time-varying observation noise, the method of Variational Bayesian learning is introduced into the SLAM framework. Firstly, the VB method is used to estimate the joint posterior probability of the AUV path and observation noise. The Inverse-Gamma distribution is used to model the observation noise and real-time noise parameters estimation is performed to improve the AUV localization accuracy. Secondly, VB is reused to estimate the noise parameters in the feature update stage to enhance the performance of the feature estimation. The proposed algorithms are first validated in an open-source simulation environment. Then, an AUV SLAM system based on the Inertial Navigation System (INS), Doppler Velocity Log (DVL), and single-beam Sonar are also built to verify the effectiveness of the proposed algorithms in the marine environment. The accuracy of the proposed methods can reach 0.742% and 0.776% of the range, respectively, which is much better than 1.825% and 1.397% of the traditional methods.
- 20 October 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 72 (3) , 2878-2890
An Integrated Navigation Method for Small-Sized AUV in Shallow-Sea Applications
The navigation and localization of Autonomous Underwater Vehicle (AUV) face many challenges. In surface missions, AUV typically relies only on Global Positioning System (GPS) to perform navigation. However, the GPS data output frequency is low, usually only 1 Hz. Meanwhile, the navigation system is susceptible to interference from GPS measurement outliers. For underwater missions, without the assistance of additional underwater absolute positioning measurements, the unknown errors introduced by the navigation system modeling process and state estimation process will affect the navigation accuracy. Therefore, in this paper, we propose an integrated navigation method for small-sized AUV in shallow-sea applications. Firstly, a robust GPS/Attitude and Heading Reference System (AHRS)/Doppler Velocity Log (DVL) navigation method for surface missions is proposed to maintain high-frequency output and robustness. Secondly, a Multiple Model (MM) integrated navigation method is proposed for AUV underwater missions. An AHRS/DVL/NavNet sub-model is proposed to perform measurement correction to correct the unknown errors. At the same time, an adaptive AHRS/DVL sub-model based on Variational Bayesian (VB) is presented to estimate the time-varying measurement noise covariance to improve the state estimation accuracy without the extra external observations. Two sub-models run parallel based on the Interacting Multiple Model (IMM). The effectiveness of the proposed integrated navigation method is verified by actual sea trial data from Sailfish 210 AUV.
- 20 October 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 72 (3) , 2878-2890
An Integrated Navigation Method for Small-Sized AUV in Shallow-Sea Applications
The navigation and localization of Autonomous Underwater Vehicle (AUV) face many challenges. In surface missions, AUV typically relies only on Global Positioning System (GPS) to perform navigation. However, the GPS data output frequency is low, usually only 1 Hz. Meanwhile, the navigation system is susceptible to interference from GPS measurement outliers. For underwater missions, without the assistance of additional underwater absolute positioning measurements, the unknown errors introduced by the navigation system modeling process and state estimation process will affect the navigation accuracy. Therefore, in this paper, we propose an integrated navigation method for small-sized AUV in shallow-sea applications. Firstly, a robust GPS/Attitude and Heading Reference System (AHRS)/Doppler Velocity Log (DVL) navigation method for surface missions is proposed to maintain high-frequency output and robustness. Secondly, a Multiple Model (MM) integrated navigation method is proposed for AUV underwater missions. An AHRS/DVL/NavNet sub-model is proposed to perform measurement correction to correct the unknown errors. At the same time, an adaptive AHRS/DVL sub-model based on Variational Bayesian (VB) is presented to estimate the time-varying measurement noise covariance to improve the state estimation accuracy without the extra external observations. Two sub-models run parallel based on the Interacting Multiple Model (IMM). The effectiveness of the proposed integrated navigation method is verified by actual sea trial data from Sailfish 210 AUV.
- 14 October 2021
- Published by SPIE-Intl Soc Optical Eng
- Vol. 11930, 150-196
Infrared image enhancement algorithm based on seagull optimized Otsu and BEEPS algorithm in NSST domain
Aiming at the problems of fuzzy, noisy, and low contrast of infrared images in detection of power equipment. This paper designs an improved Otsu segmentation threshold based on seagull optimization and BEEPS filter algorithm in NSST domain. First, the original infrared image is decomposed by NSST to high and low frequency components. Low-frequency component is divided into two parts, the foreground and background, and the enhancement processing is performed separately to each part. BEEPS algorithm is used in this paper for high-frequency component processing. Finally, the processed low-frequency components and high-frequency components are subjected to NSST inverse transformation to obtain the final enhanced image. The algorithm in this paper is compared with the other three algorithms to verify its superiority: it has improved the accuracy of infrared image threshold segmentation and strengthened the depth of field, and increase the brightness of power equipment. The overall contrast of the image is enhanced and the noise part is also effectively filtered out, improving the overall visual effect of the image which is conducive to use thermal effects to determine the operating status of power equipment and the detection and fault location of thermal fault detection.
- 14 October 2021
- Published by SPIE-Intl Soc Optical Eng
- Vol. 11930, 150-196
Infrared image enhancement algorithm based on seagull optimized Otsu and BEEPS algorithm in NSST domain
Aiming at the problems of fuzzy, noisy, and low contrast of infrared images in detection of power equipment. This paper designs an improved Otsu segmentation threshold based on seagull optimization and BEEPS filter algorithm in NSST domain. First, the original infrared image is decomposed by NSST to high and low frequency components. Low-frequency component is divided into two parts, the foreground and background, and the enhancement processing is performed separately to each part. BEEPS algorithm is used in this paper for high-frequency component processing. Finally, the processed low-frequency components and high-frequency components are subjected to NSST inverse transformation to obtain the final enhanced image. The algorithm in this paper is compared with the other three algorithms to verify its superiority: it has improved the accuracy of infrared image threshold segmentation and strengthened the depth of field, and increase the brightness of power equipment. The overall contrast of the image is enhanced and the noise part is also effectively filtered out, improving the overall visual effect of the image which is conducive to use thermal effects to determine the operating status of power equipment and the detection and fault location of thermal fault detection.
- 14 October 2021
- Published by SPIE-Intl Soc Optical Eng
- Vol. 11930, 149-187
Applications of NSST enhancement based on adaptive longicorn optimized segmentation and improved modulus in infrared image of electrical equipment
The normal state of power equipment is directly related to the operation of the system. At present, the most widely used method is to use infrared image to implement real-time monitoring of the operation status of power equipment. In order to solve the problems of noise, blur and low contrast in infrared detection, an infrared image NSST enhancement algorithm based on adaptive segmentation and improved fuzzy enhancement is designed in this paper. The original infrared image is transformed into high-frequency and low-frequency components in NSST domain. Then, the high-frequency component with noise is denoised by the spatial adaptive noise smoothing algorithm, and the improved fuzzy enhancement is used.The low-frequency component with the main body of power equipment is segmented into background and foreground parts by adaptive Longhorn, and then enhanced separately. Finally, the enhanced high-frequency and low-frequency components are inversely transformed by NSST to form the final enhanced image. Compared with the other three traditional algorithms, this algorithm has the advantages: it can not only filter the infrared image noise of power equipment effectively, but also improve the infrared image contrast, making the infrared image conform to the human visual effect, and it is easier for the human eye to recognize the fault. It is very helpful to detect and locate the thermal fault of power equipment.
- 14 October 2021
- Published by SPIE-Intl Soc Optical Eng
- Vol. 11930, 149-187
Applications of NSST enhancement based on adaptive longicorn optimized segmentation and improved modulus in infrared image of electrical equipment
The normal state of power equipment is directly related to the operation of the system. At present, the most widely used method is to use infrared image to implement real-time monitoring of the operation status of power equipment. In order to solve the problems of noise, blur and low contrast in infrared detection, an infrared image NSST enhancement algorithm based on adaptive segmentation and improved fuzzy enhancement is designed in this paper. The original infrared image is transformed into high-frequency and low-frequency components in NSST domain. Then, the high-frequency component with noise is denoised by the spatial adaptive noise smoothing algorithm, and the improved fuzzy enhancement is used.The low-frequency component with the main body of power equipment is segmented into background and foreground parts by adaptive Longhorn, and then enhanced separately. Finally, the enhanced high-frequency and low-frequency components are inversely transformed by NSST to form the final enhanced image. Compared with the other three traditional algorithms, this algorithm has the advantages: it can not only filter the infrared image noise of power equipment effectively, but also improve the infrared image contrast, making the infrared image conform to the human visual effect, and it is easier for the human eye to recognize the fault. It is very helpful to detect and locate the thermal fault of power equipment.
- 2 November 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 70 (00189456) , 1-13
Cruise Speed Estimation Strategy Based on Multiple Fuzzy Logic and Extended State Observer for Low-Cost AUV
This paper presents an effective cruise speed estimation strategy for low-cost autonomous underwater vehicle under the impact of complex ocean currents and the absence of critical navigation sensors. The estimation strategy is based on the motor feedback information, including motor current and propeller speed. The cruise speed estimation strategy requires a rough speed estimation and minor adjustments to ensure that the estimated speed is valuable. In this paper, the multiple fuzzy logic is applied to generate the mapping between the motor current deviation, propeller speed and cruise speed deviation. Afterwards, recursive least square is introduced to fine-tune the rough speed based on the reference propeller torque estimated by extended state observer. The experimental results demonstrate that the estimated speed of the proposed method can reach, on average, 96% of the Doppler velocity log measurement accuracy under the ocean currents conditions. The experimental results also illustrate that the method proposed in this paper has strong robustness.
- 2 November 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 70 (00189456) , 1-13
Cruise Speed Estimation Strategy Based on Multiple Fuzzy Logic and Extended State Observer for Low-Cost AUV
This paper presents an effective cruise speed estimation strategy for low-cost autonomous underwater vehicle under the impact of complex ocean currents and the absence of critical navigation sensors. The estimation strategy is based on the motor feedback information, including motor current and propeller speed. The cruise speed estimation strategy requires a rough speed estimation and minor adjustments to ensure that the estimated speed is valuable. In this paper, the multiple fuzzy logic is applied to generate the mapping between the motor current deviation, propeller speed and cruise speed deviation. Afterwards, recursive least square is introduced to fine-tune the rough speed based on the reference propeller torque estimated by extended state observer. The experimental results demonstrate that the estimated speed of the proposed method can reach, on average, 96% of the Doppler velocity log measurement accuracy under the ocean currents conditions. The experimental results also illustrate that the method proposed in this paper has strong robustness.