A5032200261- Underwater Vehicles and Communication Systems
- Adaptive Control of Nonlinear Systems
- Robotic Path Planning Algorithms
- Advanced Vision and Imaging
- Robotics and Sensor-Based Localization
- Distributed Control Multi-Agent Systems
- Control and Dynamics of Mobile Robots
- Advanced Control Systems Optimization
- Advanced Measurement and Metrology Techniques
- Robotic Mechanisms and Dynamics
- Manufacturing Process and Optimization
- Industrial Technology and Control Systems
- Fault Detection and Control Systems
- Soft Robotics and Applications
- Iterative Learning Control Systems
- Advanced Algorithms and Applications
- Stability and Control of Uncertain Systems
- Robot Manipulation and Learning
- Fluid Dynamics and Vibration Analysis
- Thermography and Photoacoustic Techniques
- Teleoperation and Haptic Systems
- Image Enhancement Techniques
- Vibration Control and Rheological Fluids
- Inertial Sensor and Navigation
- Photoacoustic and Ultrasonic Imaging
Northwestern Polytechnical University
2016-2025
Nanjing University
2025
Northwestern Polytechnic University
2008-2024
Genertec Shenyang Machine Tool Co., Ltd. (China)
2023
Beijing University of Posts and Telecommunications
2022
Shaanxi University of Technology
2022
Beijing Normal University
2022
Guangdong University of Technology
2005-2020
Tsinghua University
2020
Qingdao University of Science and Technology
2020
This paper proposes a hierarchical image-based visual servoing (IBVS) strategy for dynamic positioning of fully actuated underwater vehicle. In the kinematic loop, desired velocity is generated by nonlinear model predictive controller, which optimizes cost function predicted image trajectories under constraints visibility and velocity. A reference model, representing closed-loop vehicle dynamics, integrated with an IBVS to predict future trajectories. tracking neural-network-based adaptive...
The event-triggered control is a promising solution to cyber-physical systems, such as networked multiagent and large-scale intelligent systems. In this paper, we propose an model predictive (MPC) scheme for constrained continuous-time nonlinear systems with bounded disturbances. First, time-varying tightened state constraint computed achieve robust satisfaction, scheduling strategy designed in the framework of dual-mode MPC. Second, sufficient conditions ensuring feasibility closed-loop...
Side-scan sonar (SSS) images present unique challenges to computer vision due their lower resolution, smaller targets, and fewer features. Although the mainstream backbone networks have shown promising results on traditional tasks, they utilize convolution reduce dimensionality of feature maps, which may cause information loss for small targets decrease performance in SSS images. To address this problem, based yolov8 network, we proposed a new underwater target detection model upsampling...
This paper presents a fault-tolerant model predictive control approach for cross-rudder autonomous underwater vehicles to achieve heading control, considering rudder stuck faults and unknown disturbances. Specifically, additive in the rudders are addressed, an active strategy is employed. Fault models of have been established develop method. In controller design, complete failure incorporated ensure angle vehicle faulty conditions. Furthermore, fault term decoupled from input, along with...
<title>Abstract</title> Large-scale 3D photoacoustic (PA) imaging has become increasingly important for both clinical and pre-clinical applications. Limited by cost system complexity, only systems with sparsely-distributed sensors can be widely implemented, which desires advanced reconstruction algorithms to reduce artifacts. However, high computing memory time consumption of traditional iterative (IR) is practically unacceptable large-scale PA imaging. Here, we propose a point cloud-based...
Autonomous underwater vehicles (AUVs) are an important equipment for ocean investigation. Actuator fault diagnosis is essential to ensure the sailing safety of AUVs. However, lack failure data training due unknown environments and unpredictable occurrences challenging diagnosis. In this paper, a meta-self-attention multi-scale convolution neural network (MSAMS–CNN) proposed actuator Specifically, two-dimensional spectrogram vibration signals obtained by sensor used as network’s inputs. The...
Purpose The aim of the paper is to provide an economically viable solution for blade repair process. There a continual increase in market, which requires increased level specialised technology reduce cost and productivity Design/methodology/approach – This introduces aerospace component defects be repaired. Current technologies including building‐up machining are reviewed. Through analysis these available technologies, this proposes integrated strategy through information integration...
This paper presents a sliding-mode observer-based model predictive control (SMO-MPC) strategy for image-based visual servoing (IBVS) of fully-actuated underwater vehicles subject to field view and actuator constraints uncertainties. In the proposed SMO-MPC controller, system approximate vehicle are used predict future trajectories from current states driven by input candidates over certain horizon. With consideration uncertainties, including external disturbances unknown dynamic parameters,...
Abstract This article proposes an unscented Kalman filter‐based visual docking controller for underactuated underwater vehicles using a position‐based servoing (PBVS) approach. The relative pose of vehicle with respect to moving station is estimated by filter the measurements multiple point features installed on station. Based pose, Euler angles commands are designed via integral cross‐tracking method drive move along desired path. Then, adaptive neural network (NN) track yaw and pitch...
Due to the complexity and diversity of underwater environments, traditional object detection algorithms face challenges in maintaining robustness accuracy when applied underwater. This paper proposes an algorithm based on improved YOLOv8 model. First, introduction CIB building blocks into backbone network, along with optimization C2f structure incorporation large-kernel depthwise convolutions, effectively enhances model’s receptive field. improvement increases capability detecting...
Rapid developments and improvements in advanced manufacturing technology mean that the inspection process plays an important role quality control of manufactured products. The Geometric Dimension Tolerance (GD&T) is used industries to examine conformity parts with part specification defined at design stage. However, as modern trend towards producing low-volume, high-variety, high-complexity, large products, current techniques show difficulties coping this. This especially true automatically...