| Abstract Scope |
Large-area additive manufacturing (LAAM) is advancing toward intelligent, self-correcting processes enabled by integrated sensing and data analytics. In this work, laser profilometry and thermal imaging are deployed to monitor bead geometry, deposition accuracy, and thermal signatures in real time. Closed-loop algorithms, including PCA-based anomaly detection and physics-informed parametric models, analyze process data and dynamically adjust extruder speed, gantry motion, and inter-layer cooling to maintain print quality. Recent developments include the deployment of a new human–machine interface enabling automated command of printing parameters, demonstration of anomaly detection within and across builds, and identification of transferability challenges when sensors are repositioned. Hardware refinements and data augmentation strategies are explored to overcome these limitations, while model-based approaches provide interpretable metrics for robust control. These results highlight the critical role of integrated sensing and analytics in scaling LAAM toward reliable, repeatable, and industrially relevant applications. |