Title: Implementing In-Process Inspection Protocols for Automated Fiber Placement of Composites Structures

Authors: Waruna P. Seneviratne, Tharaka Nandakumara, and Aaron Jones

DOI: 10.33599/nasampe/c.25.189

Abstract: Automated Fiber Placement (AFP) is a pivotal technology in high-rate aerospace composite manufacturing. While AFP enables precision material deposition, current industry practices rely heavily on labor-intensive manual inspections that disrupt production and limit throughput. To address these inefficiencies, a novel in-process inspection system, known as the In-Process AFP Manufacturing Inspection System (IAMIS), was developed. This system integrates advanced sensor hardware with machine learning (ML) algorithms to provide detection, classification, quantification, triangulation, and real-time response (D-C-Q-T-R) during AFP. Comparative round-robin studies demonstrated IAMIS's superior performance in identifying and categorizing a range of defect types, with higher precision and sensitivity than conventional inspection systems. This paper details the system architecture, ML framework, and validation methodology, and highlights the role of AI in optimizing AFP for improved quality assurance and production efficiency. The integration of IAMIS within AFP processes offers a transformative shift from traditional, labor-intensive inspections to a fully automated, intelligent inspection workflow that ensures compliance with stringent aerospace manufacturing standards while significantly reducing cycle time and production cost.

References: 1. Rudberg, T.; Nielson, J.; Henscheid, M.; Cemenska, J. Improving AFP Cell Performance. SAE Int. J. Aerosp. 2014, 7, 317–321, doi:10.4271/2014-01-2272. 2. Seneviratne, W., Tomblin, J., and Palliyaguru, U., “Machine-Learning for Automated Fiber Placement for manufacturing Efficiency and Process Optimization,” Society for the Advancement of Material and Process Engineering (SAMPE), May 2021. 3. Seneviratne, W., Tomblin, J., Nandakumara, T., and Jones, A., “Automated In-Process AFP Manufacturing Inspection System (IAMIS) Development and Round-Robin Evaluation,” AFRL Manufacturing for Affordable Sustainable Composites, (AFRL Report Under Review). 4. Bishop C.M., Pattern Recognition and Machine Learning. New York: Springer, Aug. 2006. ISBN 978-0-387-31073-2.

Conference: CAMX 2025

Publication Date: 2025/09/08

SKU: 189

Pages: 15

Price: $30.00