Chemical Servoing with Hexapod Robot for Gas Source Localization and YOLO-based Visual Defect Detection
Article : Articles dans des revues internationales ou nationales avec comité de lecture
Robotic systems have the ability to perform industrial inspection tasks through
the integration of visual and chemical servoing functions. A commonly used system
is the image-based servoing system. This study presents an innovative approach
that merges chemical servoing with visual defect detection in industrial settings
using a hexapod robot. The main goal is to develop a hexapod robot capable of
detecting and approaching a gas source through chemical servoing, and subsequently
analyzing the source using a pretrained YOLOv5m model, a one-stage CNN detector
architecture. The effectiveness of the proposed methodology is showcased through
simulations that compare different chemical servoing algorithms and image defect
detection techniques for gas source identification. In addition, a prototype of the
hexapod robot is presented.