A5075186257- Robotics and Sensor-Based Localization
- Industrial Vision Systems and Defect Detection
- Advanced Vision and Imaging
- Image and Object Detection Techniques
- Image Processing Techniques and Applications
- Advanced Battery Technologies Research
- Transportation and Mobility Innovations
- 3D Surveying and Cultural Heritage
- Digital Media Forensic Detection
- Underwater Vehicles and Communication Systems
- Electric Vehicles and Infrastructure
- Video Surveillance and Tracking Methods
- Advanced Neural Network Applications
- Image Processing and 3D Reconstruction
- Underwater Acoustics Research
- Network Time Synchronization Technologies
- Distributed Control Multi-Agent Systems
- Stability and Control of Uncertain Systems
Huazhong University of Science and Technology
2004-2023
Ministry of Education of the People's Republic of China
2023
For the existing visual-inertial SLAM algorithm, when robot is moving at a constant speed or purely rotating and encounters scenes with insufficient visual features, problems of low accuracy poor robustness arise. Aiming to solve inertial tightly coupled vision-IMU-2D lidar odometry (VILO) algorithm proposed. Firstly, low-cost 2D observations are fused in manner. Secondly, model used derive Jacobian matrix residual respect state variable be estimated, constraint equation constructed....
This paper studies the networked control systems with multiple-packet transmission. We consider time-driven sensors, event-driven controller and actuators in MIMO NCS. Each sensor node transmits sampled data one packet. present model of NCS The stability system is also discussed.
In industrial production processes, defect inspection plays an important role in reducing the occurrence of failures and improving efficiency. Data-driven algorithms represented by deep learning have made great progress recent years, but need to face problems small quantity poor quality datasets when applied inspection. This paper proposes a layer mask blending-based generative adversarial network (LMBGAN) optimizes training process generate high-quality surface samples. LMBGAN generates...
Semantic Segmentation is the foundation of scene understanding and automatic driving tasks. One challenges semantic segmentation reduction feature resolution as network goes deep. In this paper, low-resolution features are integrated into high-resolution progressively, enabling prediction. Currently, incorporating auxiliary depth information a framework has proven to be helpful improve accuracy. However, additional process data immensely increases computational complexity, resulting in...
The key for bulldozers to realize automatic operation in mine scenes is whether they can accurately identify and segment the retaining walls, however, because point cloud dataset of site too few, resulting deep learning-based identification segmentation algorithm cannot be applied this scene. At same time, rugged road surface scene, existence a lot dust, numerous disturbances, wall features are not obvious other problems, algorithms proposed by previous scholars perfect enough solve problem....
In the industrial production environment, scarcity of defect samples and high labor cost labeling make supervised machine learning models difficult to implement. addition, detection using deep usually requires computation advanced hardware platform, which is not suitable for running on embedded platforms. To solve this problem, paper proposes an unsupervised fabric method based lightweight convolutional denoising auto-encoder(LCDAE). The LCDAE model constructed by bottleneck layer in...
Electric vehicles (EVs) are some of the major consumers renewable energy generation in recent years. However, uncertainties associated with photovoltaic (PV) output and EV users' charging behaviors have imposed great challenges on PV power consumption. This study presents a real-time strategy based deep reinforcement learning (DRL) to reduce curtailment costs users. A mathematical model is constructed describe control process EVs, which outputs considered. problem then formulated as Markov...