Highlighted research posts
Quantitative control of flow rate using deep learning
In this work, we harness commonly available manufacturing process metadata alongside video to train deep learning regression models. These models we then apply to closed-loop control and few-shot error detection in extrusion 3D printing.
Detecting and correcting 3D printing errors on the fly
We developed a generalisable deep learning model that can detect and correct a wide variety of different 3D printing errors in real time, which can be easily added to new or existing machines to enhance their capabilities. 3D printers using the algorithm could also learn how to print new materials by themselves.
Automated recognition and correction of warp
Warp deformation is one of the most commonly encountered errors in additive manufacturing. We applied a combination of deep learning and expert heuristics to create the first system to autonomously recognise and correct warp both in situ and for future prints.