This work presents a novel approach to improving the robotic quality inspection of the spray coating process in the aerospace industry by integrating computer vision with robotic systems. While spray coating is essential for providing protective and aesthetic coatings, due to the challenges and complexity of the aerospace industry, it frequently encounters issues such as incomplete coverage, paint defects, and surface imperfections, which can compromise quality and increase the need for rework. To address this, a methodology that utilises a cutting-edge computer vision technique based on YOLOv10 for realtime defect localisation is proposed, targeting issues such as uneven thickness and missed areas. Once the camera is calibrated, the results of defect localisation achieve a multi-class mean Average Precision of 99%. Furthermore, this work presents a framework that demonstrates how positional information and classification results can be utilised to automatically generate path planning and control actions for an intelligent spray coating system. This innovation advances the state of knowledge in the field, which has previously relied only on image classification.
Vision-based defect localisation and automated planning for robotic spray coating systems / Forni, Tommaso; Mattera, Giulio; Vozza, Mario; Nele, Luigi. - In: JOURNAL OF THE BRAZILIAN SOCIETY OF MECHANICAL SCIENCES AND ENGINEERING. - ISSN 1678-5878. - 47:9(2025). [10.1007/s40430-025-05746-z]
Vision-based defect localisation and automated planning for robotic spray coating systems
Forni, Tommaso;Vozza, Mario;
2025
Abstract
This work presents a novel approach to improving the robotic quality inspection of the spray coating process in the aerospace industry by integrating computer vision with robotic systems. While spray coating is essential for providing protective and aesthetic coatings, due to the challenges and complexity of the aerospace industry, it frequently encounters issues such as incomplete coverage, paint defects, and surface imperfections, which can compromise quality and increase the need for rework. To address this, a methodology that utilises a cutting-edge computer vision technique based on YOLOv10 for realtime defect localisation is proposed, targeting issues such as uneven thickness and missed areas. Once the camera is calibrated, the results of defect localisation achieve a multi-class mean Average Precision of 99%. Furthermore, this work presents a framework that demonstrates how positional information and classification results can be utilised to automatically generate path planning and control actions for an intelligent spray coating system. This innovation advances the state of knowledge in the field, which has previously relied only on image classification.File | Dimensione | Formato | |
---|---|---|---|
s40430-025-05746-z.pdf
accesso riservato
Tipologia:
2a Post-print versione editoriale / Version of Record
Licenza:
Non Pubblico - Accesso privato/ristretto
Dimensione
2.5 MB
Formato
Adobe PDF
|
2.5 MB | Adobe PDF | Visualizza/Apri Richiedi una copia |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/11583/3001673