A deep learning solution for real-time quality assessment and control in additive manufacturing using point cloud data

This work presents an in-situ quality assessment and improvement technique using point cloud AI for data processing smart decision making in Additive Manufacturing (AM) fabrication to improve the accuracy of fabricated artifacts. The top surface point-cloud containing geometry information is pre-processed passed improved deep Hybrid Convolutional Auto-Encoder decoder (HCAE) model used statistically describe artifact's quality. HCAE’s output comprised 9 × segments, each including four channels with segment's probability contain one labels, Under-printed, Normally-printed, Over-printed, or Empty region. structure plays a significant role command generation process optimization. repeatability were measured by multi-label multi-output metric developed this study. results are perform real-time adjustment manipulating future layer's through G-code modification. By adjusting machine's print speed feed-rate, controller exploits subsequent layer’s deposition, grid-by-grid. algorithm then tested two defective plans: severe under-extrusion over-extrusion conditions. Both test artifacts' advanced significantly converged acceptable state iterations.