A5075217125- Underwater Vehicles and Communication Systems
- Target Tracking and Data Fusion in Sensor Networks
- Adaptive Control of Nonlinear Systems
- Maritime Navigation and Safety
- Underwater Acoustics Research
- Robotics and Sensor-Based Localization
- Robotic Path Planning Algorithms
- Water Quality Monitoring Technologies
- Inertial Sensor and Navigation
- Fault Detection and Control Systems
- Advanced Control Systems Optimization
- Fluid Dynamics Simulations and Interactions
- Aerospace Engineering and Energy Systems
- Control and Dynamics of Mobile Robots
- Ship Hydrodynamics and Maneuverability
- Image Enhancement Techniques
- Machine Fault Diagnosis Techniques
- Advanced Neural Network Applications
- Advanced Computational Techniques and Applications
- Fish biology, ecology, and behavior
- Hydraulic and Pneumatic Systems
- Neural Networks and Applications
- Soft Robotics and Applications
- Advanced Image Processing Techniques
- Biomimetic flight and propulsion mechanisms
Zhejiang Ocean University
2016-2025
Zhejiang University
2016-2025
Shenyang Institute of Automation
2010-2024
Chinese Academy of Sciences
2010-2024
Huzhou University
2022-2024
City University of Hong Kong, Shenzhen Research Institute
2021-2023
Ministry of Education
2022
Zhoushan Hospital
2021
University of Plymouth
2017
Tianjin University of Technology and Education
2015
A novel integrated navigation algorithm, multi-model EKF (Extended Kalman Filter) is presented in this paper for the deep water autonomous underwater vehicle. When a vehicle performing tasks sea, error will accumulate over time, if it relies solely on its own inertial system. In order to get more accurate position online, an system constructed by adding acoustic And because difficult establish kinematic model and measurement accurately environment, we propose Multi-model estimate errors of...
Small Autonomous Surface/Underwater Vehicles (S-ASUVs) are gradually attracting attention from related fields due to their small size, low energy consumption, and flexible motion. Existing dynamic positioning (DP) control approaches suffer chronic restrictions that hinder adaptability varying practical conditions, rendering performance poor. A new three-dimensional (3D) method for S-ASUVs is proposed tackle this issue. Firstly, a model the DP problem considering thrust allocation was...
Regular surface-maintenances are necessary for high structures to increase service life. The traditional manual operation has shortcomings like limited maneuverability, poor operating quality, low efficiency, and risk of physical harm, which makes it urgent develop wall-climbing robot carrying out with cost, good protection operators. In this article, we have developed a wheeled that uses permanent magnet adhesion system climb on large steel surfaces. Wheel traction avoid slippage is...
The dataset contains 1225 data samples for 5 fault types (labels). We divided the into training set and test through random stratified sampling. accounted 20% of total dataset. Our experimental subject is 'Haizhe', which a small quadrotor AUV developed in laboratory. For each type, 'Haizhe' was tested several times. time, ran same program sailed underwater 10-20 s to ensure that state long enough. recorded were then used as sample, corresponding type true label sample. validate model-free...
Ocean science, which delves into the oceans that are reservoirs of life and biodiversity, is great significance given cover over 70% our planet's surface. Recently, advances in Large Language Models (LLMs) have transformed paradigm science. Despite success other domains, current LLMs often fall short catering to needs domain experts like oceanographers, potential for ocean science under-explored. The intrinsic reason may be immense intricate nature data as well necessity higher granularity...
The fixed-time strategy is crucial in traffic signal control which applies plans to different time periods of the day. One critical step determine optimal breakpoints divide one day into with homogeneous flow. Most existing methods are based on k-means clustering algorithm and have select number clusters. Since direct time-incorporated will result noncontiguous periods, several adjustments needed including further partitioning re-clustering via empirically adjustment merges short adjacent...
In this paper, we develop a new classification method for manifold-valued data in the framework of probabilistic learning vector quantization. many scenarios, can be naturally represented by symmetric positive definite matrices, which are inherently points that live on curved Riemannian manifold. Due to non-Euclidean geometry manifolds, traditional Euclidean machine algorithms yield poor results such data. generalize quantization algorithm living manifold matrices equipped with natural...
A solution to the path following problem for underactuated autonomous vehicles in presence of possibly large modeling parametric uncertainty is proposed. For a general class moving 2D space, we demonstrated control law based on multiple variable sliding mode that yields global boundedness and convergence position tracking error small neighborhood robustness uncertainty. An integration element added into “tanh” function traditional control. We illustrated our results context vehicle...
In this article, an autonomous robotic fish is designed for underwater operations like object detection and tracking along with collision avoidance. The computer-aided design model prototype using the Solid Works® software to export stereolithography (STL) file MakerBot, a 3D printer, manufacture parts of polylactic acid thermoplastic polymer. precise maneuverability achieved by propulsion caudal fin. oscillation fin controlled servomotor. A combination visual ultrasonic sensors used track...