A5103039547- Robotic Path Planning Algorithms
- Robotics and Sensor-Based Localization
- Advanced Image and Video Retrieval Techniques
- 3D Surveying and Cultural Heritage
- Thermography and Photoacoustic Techniques
- Infrared Target Detection Methodologies
- Inertial Sensor and Navigation
- Remote Sensing and LiDAR Applications
Nanjing University of Posts and Telecommunications
2016-2024
Southern University of Science and Technology
2021-2022
Mobile robots need reliable maps for autonomous operation. Traditional SLAM systems, which are mainly developed static scenes, often fail in dynamic environments with moving objects present the scene. Learning based systems suffer from insufficient or inaccurate identification of feature points. This paper proposes a novel real-time RGB-D system, is targeted environments, can further enhance detection and removal. done by fusing panoptic segmentation geometry information. The system includes...
Generating high-fidelity, controllable, and annotated training data is critical for autonomous driving. Existing methods typically generate a single form directly from coarse scene layout, which not only fails to output rich forms required diverse downstream tasks but also struggles model the direct layout-to-data distribution. In this paper, we introduce UniScene, first unified framework generating three key - semantic occupancy, video, LiDAR in driving scenes. UniScene employs progressive...
Robots can perform various missions in multiple changing environments. The dynamic objects have significant influence on the long-term autonomy and 3D map construction, because "ghost tracks" inevitably exist due to continuous error-accumulation of input data. So it is critical keep only static subsets exclude noisy obstacles mitigate mapping navigation. We propose a robust building method, which compares discrepancies between single scan data against map. This method focuses advantages most...