EuropaWire: Self-Guiding Laser Cutting Uses Cameras and Microphones

Researchers at Empa, SUPSI, and Swiss machine manufacturer Bystronic have developed a self-guiding laser cutting method that utilizes cameras and microphones, along with machine learning, to assess cut quality and automatically adjust machine settings. This project addresses the common challenge of laser cutting machines requiring time-consuming test series and recalibration for different alloys and material thicknesses, especially when processing thick metal.

The system aims to give the laser cutting machine "eyes" and "ears" to monitor the process autonomously. SUPSI is developing camera-based evaluation of cut quality, while Empa is creating acoustic methods that use sound recordings from microphones placed inside the machine. Researchers found that acoustic assessment can evaluate cut quality nearly as effectively as cameras but requires less expensive equipment.

To train their models, the team conducted tests using metal thicknesses of 6 mm, 10 mm, 15 mm, and 25 mm, focusing on cut-edge roughness and burr formation. They used nine microphones placed at different positions within the machine and simultaneously recorded ambient noise to train the acoustic models.

The project partners are now linking the acoustic and optical models to a central control system designed to allow the laser cutting machine to immediately optimize its own settings. This capability could eliminate the need for laborious test series by enabling automatic adjustment to the best parameters for each material. The two-year project is expected to conclude in autumn 2026, and the resulting system is intended to be retrofittable to existing machines.