AI4CADCAM: AI-based processing of CAD models for automated planning of computer-aided manufacturing

Team Members: Konstantinos Gkrispanis, Stavros Nousias, André Borrmann

Funding: Bayerischen Staatsministerium für Wirtschaft, Landesentwicklung und Energie (StMWi)

Partners: TEBIS AG , TEBIS Technische Informationssysteme AG

Running Period: 01.12.2023-31.12.2026

In the context of modern manufacturing, computer-aided manufacturing (CAM) processes are increasingly utilized for producing intricate machine components. This involves programming NC machines for precise material removal, a task that demands considerable expert knowledge for selecting appropriate manufacturing tools and defining their movements. As the complexity of parts to be processed grows, coupled with an increasing demand of CAM specialists, there arises a compelling need for automation in this planning stage.

To address this challenge, the AI4CADCAM project emerges as an innovative initiative, poised to transform CAM through the integration of artificial intelligence (AI). This project focuses on automating the planning process, which includes the intricate selection and orientation of manufacturing tools. It leverages geometry-driven generative AI methods, facilitating a seamless transition from the initiation of manufacturing processes to the derivation of process steps and parameters directly from CAD models.

AI4CADCAM is dedicated to advancing novel AI methodologies in Geometric Deep Learning and AI-driven decision making. By deploying cutting-edge approaches like Transformer-based architectures, along with Convolutional Neural Networks (CNNs), Variational Autoencoders (VAEs), and Graph Neural Networks (GNNs), the project aims to automate various facets of the CAD-CAM process. This ambition promises to achieve a level of automation in CAM that was previously unattainable, marking a significant leap forward in the field.