Research project DetectFixAM

This project aims to guarantee first-time-right quality in material extrusion-based 3D printing. It will develop a dual approach: (1) real-time monitoring using multi-modal sensors (visual, thermal, acoustic) combined with AI models to predict defects, and (2) adaptive correction strategies to restore part integrity during printing. Machine learning models will be trained on a dedicated defect database, leveraging data fusion techniques for robust anomaly detection. Corrective actions may include process adjustments, manual interventions, or automated in-situ repairs. Final part quality will be validated through non-destructive testing to ensure performance parity with defect-free prints. This approach will significantly improve reliability and reduce waste in additive manufacturing.