Research of a Foreign Objects Detection Fusion Algorithm Using Video and LIDAR in FAO System
- Resource Type
- Conference
- Authors
- Yu, Qingguang; Wang, Shi; Liu, Youqi; Zhang, Lintao; Tan, Ashton Yuxuan; Siow, Chengbo; Wang, Yujin; Cai, Guanzi
- Source
- 2022 5th International Conference on Renewable Energy and Power Engineering (REPE) Renewable Energy and Power Engineering (REPE), 2022 5th International Conference on. :73-77 Sep, 2022
- Subject
- Power, Energy and Industry Applications
Point cloud compression
Laser radar
Three-dimensional displays
Radar detection
Object detection
Reliability engineering
Safety
fully automatic operation
subway screen doors
fusion algorithms
automatic detection
- Language
- ISSN
- 2771-7011
This paper aims to research a fusion algorithm for subway screen doors to make future fully automatic operation (FAO) systems safer and more efficient. This paper adopts video with light detection and ranging (LIDAR) fusion algorithm technology, and proposes a dual-criteria artificial intelligence (AI) strategy using video image recognition and radar point cloud fusion to detect foreign objects in the gap between subway platform doors. A novel approach of cross-stacking and layered installation of sensors is innovatively proposed to realize the function of redundant detection of foreign objects in the gap. Cross-checking improves the reliability of the detection device while 2D sensors are used to achieve 3D detection effects. The developed system provides safety interlocking signals for the subway signal system, delivering alarm information to the integrated monitoring system, then pushes the smart hand ring alarm information to the on-site operators.