Industrial organizations in Germany and Canada are collaborating to provide automated control of laser material deposition (LMD) and other additive processes using artificial intelligence (AI). The partners in the Artificial Intelligence Enhancement of Process Sensing for Adaptive Laser Additive Manufacturing (AI-SLAM) project aim to develop an AI-powered expert system to automate LMD to detect quality issues swiftly, reliably, and economically. LMD is used to make precise repairs to industrial machines that receive a lot of use and require frequent overhauls. An operator will use an LMD process to apply new layers to the equipment until it is restored to its original geometric shape. The uneven wear that is typical for most parts means that layers of varying thickness have to be applied, requiring the operator to measure the thickness of the coating and readjust the process frequently. The AI-SLAM project consortium partners are jointly developing a process control software system that plant manufacturers can use to run LMD processes automatically. The system records geometries during the coating process, detects deviations from the specified contour, and readjusts process parameters, such as the feed rate, accordingly. The consortium is using AI to optimize the control parameters and increase productivity. The software analyzes a large data set and independently learns how to iteratively improve the process. As much process data as possible is collected. From this data, complex routines learn automatically how process control can be optimized to produce more with less effort. The consortium expects that high efficiency gains will be made through automated process control. Parts subject to wear, such as this stone crusher tooth with an outer diameter of about 140 mm, are restored with the LMD process. Thanks to AI, the processes for repairing irregular surfaces will be optimized. Courtesy of Apollo Machine and Welding Ltd., Canada. One of German consortium partners, the Fraunhofer Institute for Laser Technology (ILT), has met a milestone in the three-year project by commissioning the software functionality for both the scanning components and the automatic path planning. In addition to Fraunhofer ILT, the consortium partners include the software developer BCT Steuerungs-und DV-Systeme GmbH (BCT) in Germany and the National Research Council of Canada (NRC), Braintoy Inc., Apollo Machine and Welding Ltd., and McGill University in Canada. A McGill University team is engaged in the research for the project. Apollo Machine and Welding is participating in the project as an industrial service provider for LMD. Braintoy is programming the machine learning algorithms for the system. The partners are using Braintoy’s web platform, Machine Learning Operating System (mlOS), to exchange process data and implement the machine learning models. For complex geometries such as on this blade tooth, or where wear is uneven, AI-based process optimization will enable significant gains in efficiency. Courtesy of Apollo Machine and Welding Ltd., Canada. Collaboration across continents is supported through regular video meetings and jointly prepared online documents. Through virtual lab tours, the partners have become acquainted with each other’s software and hardware environments. The AI-SLAM project began in April and will run until March 2024 as part of the 3+2 Canada-Germany Collaborative Industrial Research and Development Program. The program is funded on the German side by the Federal Ministry of Education and Research and on the Canadian side by the NRC. The 3+2 Canada-Germany Collaborative Industrial Research and Development Program is focused on using AI to develop new technologies for industrial production in several industries, including automotive, metals manufacturing, materials, telecommunications, energy, mining, construction, and infrastructure management. Both countries’ innovation ecosystems benefit from a mix of industries and institutions working together to provide technical and market solutions in cutting-edge research areas, such as AI applications in manufacturing.