One of the next key evolutionary phases of additive manufacturing is the full implementation of AI, machine learning and computer vision systems into the AM process in order to catch, eliminate and even correct errors, while improving the overall build process. We know that many experts are working on this, just a couple of days we reported about the Inkbit system that spun off from MIT CSAIL. Now it’s the turn of another company that is so much based on AI that it is even part of its name: London based Ai Build.
Ai Build focuses on multi-axis robotic extrusion 3D printing of very large structures using polymers. Though not necessarily as dramatic as for aerospace or oil and gas parts, a defect in an architectural or artistic part can have costly repercussions. With the new AI implementation Production, these defects are no longer an obstacle to large-scale 3D printing. The technology automatically detects most common printing issues such as material shrinkage and warping in real time. A deep neural network trained by Ai Build increases productivity and level of automation in additive manufacturing facilities and at the same time reduces production costs and material waste.
How Ai Build implements AI for error detection
“We have a simple rule at Ai Build: if a product is faulty, we repeat production. If it has a small defect, we repeat. If we are slightly in doubt, we repeat,” explained Daghan Cam, CEO of Ai Build. “This level of perfection in additive manufacturing usually comes at a high cost, in the form of excessive labor and material waste. The milestone we are announcing today is a result of our obsession for flawless production and a major leap towards fully automated, autonomous factories of the future. We are excited to see how the productivity of our users will increase with this advanced technology”
With an embedded camera and a GPU module, the AiMaker (the company’s Autonomous Large Scale 3D Printing technology) can detect manufacturing defects by comparing its real-time camera view to previously recorded images of 3D printed parts. Users of the new AiMaker will be able to get instant notifications about the quality of their 3D printed parts through the AiSync cloud software.
“By detecting printing defects in real-time, our AI technology offers the potential of automating quality control, while at the same time acting as a form of feedback to the designer about the printability of their design and how such printing issues could be eliminated in future prints,” said Leonidas Leonidou, Ai Build’s Head of Applied Research.
This new feature of AiMaker will be presented at the ExCel London as part of the Ai Summit taking place on 12th-13th June 2019. Ai Build will start shipping its new generation 3D printers with computer vision capability in the second half of 2019.