NSF Future Manufacturing Data Challenge: A Multimodal DED Dataset for Probabilistic Local Geometry Prediction in Laser Tracks

This dataset supports the NSF Future Manufacturing Data Challenge and contains multimodal measurements for probabilistic local geometry prediction in single directed energy deposition (DED) laser tracks on SS316L substrates. The dataset includes in-situ thermal image sequences, SEM images of surrounding substrate morphology, and Bruker/Wyko full-field height maps. The common analysis window corresponds to physical part coordinates 20–100 mm. The dataset is intended for developing models that predict local track width, boundary position, contour deviation, edge roughness, or related probabilistic descriptors from process-driven and substrate-driven information.