VisionAdapt – Imaging-based chip formation detection for enhancing autonomy in high-dynamic turning processes

In turning operations, long and unbroken chips often occur, which can damage tools and workpieces, preventing autonomous manufacturing. These chip forms are influenced not only by material properties but also by the selection of process parameters and the setting angle. 3-axis simultaneous turning provides an additional degree of freedom that can be used to improve chip breaking. However, due to the complex interplay of these influencing factors and the stochastic nature of chip breakage, predictions are inherently uncertain.

Objectives

The ZDIN transfer project VisionAdapt is developing an innovative solution to this problem. First, critical chip forms are to be predicted and minimized already during tool path planning, fully leveraging the advantages of 3-axis simultaneous turning. Since these predictions involve uncertainties, AI-based image recognition is employed to detect critical chip forms. This enables adaptive, online tool path planning that can respond to critical chip forms in a model-driven manner without interrupting the process. The key advantage: instead of stopping the process or requiring manual intervention, the system can autonomously adjust the feed to produce shorter chips.

Benefits

• Process reliability – automatic response to critical chips
• Productivity – fewer tool changes, reduced downtime
• Quality – stable machining conditions
• Autonomy – a step toward self-optimizing manufacturing

Approach

To achieve this goal, three institutes are collaborating with an industrial partner: The IFW at Leibniz University Hannover, as project coordinator, develops and implements the adaptive tool path planning; the DFKI in Oldenburg designs the image processing algorithms and AI models for real-time chip detection. The industrial partner DMG MORI AG, as a leading machine tool manufacturer, contributes its expertise and will transfer the results to end users. The Institute for Mechanical Design at TU Braunschweig gathers user requirements from industrial operators, thus supporting the technology transfer.

Are you also interested in a cooperation project?

Contact Felix Zender  via email at zender@ifw.uni-hannover.de or by phone at +49 511 762 18359.