An Obstacle Perception Algorithm Based on Multi-Sensor Fusion for Autonomous-Rail Rapid Transit
This paper presents an obstacle perception algorithm based on multi-sensor fusion, aimed at addressing omissions and errors, as well as low accuracy in object detection for autonomous-rail rapid transit (ART). In a structure integrating pre-fusion and post-fusion algorithms, this approach combines t...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Office of Control and Information Technology
2024-08-01
|
| Series: | Kongzhi Yu Xinxi Jishu |
| Subjects: | |
| Online Access: | http://ctet.csrzic.com/thesisDetails#10.13889/j.issn.2096-5427.2024.04.014 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This paper presents an obstacle perception algorithm based on multi-sensor fusion, aimed at addressing omissions and errors, as well as low accuracy in object detection for autonomous-rail rapid transit (ART). In a structure integrating pre-fusion and post-fusion algorithms, this approach combines the advantages of range detection by LiDAR, category detection by cameras, and speed detection by millimeter-wave radars. The pre-fusion process, supported by partial sensors, increases detection accuracy while avoiding excessive computational loads to on-board processing systems. Meanwhile, the post-fusion algorithm introduces redundancy to maintain the detection system valid when a single algorithm is out of work, as a means to ensure operational safety. Experimental results showed that the proposed method effectively detected obstacles on the operational tracks, outperforming single-sensor detection in terms of object data, and achieving detection distances longer than 70 m. These findings demonstrate its capabilities in ensuring the safe and stable operation of autonomous-rail rapid trams. |
|---|---|
| ISSN: | 2096-5427 |