Hydrogen refuelling docking control driven by multi-scale perception and task decoupling
Abstract
This paper addresses the challenges of accurate axial force control and extended docking time during rapid docking between six-axis robotic arm-mounted hydrogen refuelling guns and hydrogen ports.A multi-scale collaborative perception-driven hydrogen port recognition method and a force/position hybrid control strategy based on task space decoupling are proposed.In the recognition stage, multi-scale collaborative perception is adopted: first, the YOLOv10 algorithm performs coarse-grained image detection to quickly locate candidate regions of the hydrogen port; then, the original image is adaptively cropped based on candidate regions to limit the semantic segmentation scope to the target region, enabling fine-grained recognition.
Fields of Study & Subjects
Robot Manipulation and Learning
Teleoperation and Haptic Systems
Advanced Neural Network Applications
Publication details
- Journal / source
- International Journal of Information and Communication Technology
- Volume
- 27
- Issue
- 25
- Pages
- 42–65
- Publication date
- 1 January 2026
- Language
- EN