A5102719042- Advanced Image and Video Retrieval Techniques
- Robotics and Sensor-Based Localization
- HVDC Systems and Fault Protection
- Microgrid Control and Optimization
- Advanced Vision and Imaging
- Smart Grid Energy Management
- Optimal Power Flow Distribution
- Visual Attention and Saliency Detection
University of Electronic Science and Technology of China
2023-2024
McGill University
2017
Visual Simultaneous Localization and Mapping (VSLAM) has undergone gradual development found widespread application. However, existing VSLAM systems predominantly rely on static environment assumptions, leading to diminished robustness localization accuracy in the presence of dynamic elements. Previous research primarily employed geometric semantic constraints address regions scene. Nevertheless, their efficacy is limited complex scenarios involving non-rigid objects, non-predefined motion...
This study shows that with fast and flexible power electronic converter interfaces, micro‐grids can quickly re‐energise blacked‐out transmission lines without overvoltage. After the are re‐energised, speed up (i) reconnections of generators loads; (ii) re‐synchronisation severed line sections back to original grid. Micro‐grids have active energise auxiliary motors large thermal plants in black start. Faster restoration is limited by steam‐turbine because they cannot be speeded beyond...
Simultaneous Localization and Mapping (SLAM) has become a critical technology for intelligent transportation systems autonomous robots is widely used in driving. However, traditional manual feature-based methods challenging lighting environments make it difficult to ensure robustness accuracy. Some deep learning-based show potential but still have significant drawbacks. To address this problem, we propose novel hybrid system visual SLAM based on the LightGlue learning network. It uses local...
Autonomous robots sometimes need to operate in low-light environments, which can present significant challenges for simultaneous localization and mapping (SLAM) technology. Traditional SLAM systems based on visible-light cameras struggle function effectively under conditions, leading researchers explore alternative thermal sensors. However, sensors have limited feature information due their low resolution. To fully leverage the from both visible images while minimizing hardware dependency,...