Yufei Xu

The simultaneous localization and mapping (SLAM) technique provides long-term near-seafloor navigation for autonomous underwater vehicles (AUVs). However, the stability of descriptors generated by interest point detectors remains a challenge in the hydrothermal environment. This paper proposes a dual-branch feature generalization method, incorporating volumetric density and color distribution for enhanced robustness. The method utilizes shared descriptors and a feature confidence mechanism, combining neural radiance fields with Gaussian splatting models, ensuring fast and accurate feature generalization. The proposed approach improves recall while maintaining matching accuracy, ensuring stability and robustness in feature matching. This method achieves stable and reliable feature matching in a simulated hydrothermal environment.

The aim of this study was to address the trajectory tracking control problem of benthic autonomous underwater vehicles (AUVs) subjected to the unwinding phenomenon, lumped uncertainties, and communication bandwidth limitations. The hydrodynamic model of the AUV expressed using the Euler angle exhibits an unwinding phenomenon, and the attitude angle cannot be uniquely represented. Hence, the attitude dynamics model is restructured using a rotation matrix to solve this problem. Because of the growing complexity of the controller design for employing a rotation matrix representation for the attitude model, it is essential to adopt an alternative error format that facilitates simplified calculations. Subsequently, a new singularity-free sliding mode control law is proposed to satisfy finite-time convergence. Furthermore, a bounded adaptive law is introduced to enhance the robustness under lumped uncertainties. Additionally, an event-triggering mechanism is implemented for the control laws to alleviate the communication burden on the AUV control system. Theoretical analysis and simulations demonstrate the effectiveness and superiority of the proposed method.

The simultaneous localization and mapping (SLAM) technique provides long-term near-seafloor navigation for autonomous underwater vehicles (AUVs). However, the stability of the interest point detector (IPD) remains challenging in the seafloor environment. This paper proposes an optical–acoustic fusion interest point detector (OAF-IPD) using a monocular camera and forward-looking sonar. Unlike the artificial feature detectors most underwater IPDs adopt, a deep neural network model based on unsupervised interest point detector (UnsuperPoint) was built to reach stronger environmental adaption. First, a feature fusion module based on feature pyramid networks (FPNs) and a depth module were integrated into the system to ensure a uniform distribution of interest points in depth for improved localization accuracy. Second, a self-supervised training procedure was developed to adapt the OAF-IPD for unsupervised training. This procedure included an auto-encoder framework for the sonar data encoder, a ground truth depth generation framework for the depth module, and optical–acoustic mutual supervision for the fuse module training. Third, a non-rigid feature filter was implemented in the camera data encoder to mitigate the interference from non-rigid structural objects, such as smoke emitted from active vents in hydrothermal areas. Evaluations were conducted using open-source datasets as well as a dataset captured by the research team of this paper from pool experiments to prove the robustness and accuracy of the newly proposed method.

The aim of this study is to address the trajectory-tracking control problem of benthic autonomous underwater vehicles (AUVs) subjected to model uncertainties and extra disturbance. In order to estimate lumped uncertainties and reconstruction speed information, this paper designs a robust prescribed-time extended state observer (RPTESO), and its prescribed time can be directly designed as an explicit parameter, without relying on the initial state of the system and complex parameter settings. In addition, an adaptive law is designed to improve the robustness of RPTSEO and reduce overshoot on the premise of ensuring convergence speed. Then, a non-singular robust practical predefined-time sliding mode control (RPPSMC) considering the hydrodynamic characteristics of AUV is designed, and the predefined time can be directly set by an explicit parameter. The RPPSMC is designed based on the lumped uncertainties estimated using RPTESO, so as to improve the control accuracy of the controller in a complex environment. Theoretical analysis and simulations demonstrated the effectiveness and superiority of the proposed method.

Considering intense hydrothermal activities and rugged topography in a near-bottom environment of the trans-Atlantic geotraverse (TAG) hydrothermal mound, a small autonomous underwater vehicle (S-AUV) will suffer from time-varying disturbances, model uncertainties, actuator faults, and input saturations. To handle these issues, a fault-tolerant adaptive robust sliding mode control method is presented in this paper. Firstly, unknown disturbances, model uncertainties, and actuator faults of the S-AUV are synthesized into a lumped uncertain vector. Without requiring the upper bound and gradient of the uncertainties, a continuous adaptive finite-time extended state observer is designed to estimate the lumped uncertain vector. Then, an auxiliary dynamic system composed of continuous functions is introduced to deal with input saturations, thereby contributing to achieving fixed-time trajectory tracking control of S-AUVs. Based on a designed continuous fixed-time nonsingular fast sliding mode surface, the proposed continuous adaptive controller is chattering free. Simulated topography is built according to topographic data of the TAG mound, and a smooth trajectory model is constructed by cubic spline interpolation. Comprehensive simulations performed on an actual S-AUV model are given to validate the effectiveness and superiority of the presented algorithm.