Autonomous ultra short pulse ablation process design with Bayesian optimization

Ultra-short pulse laser ablation allows precise material removal with minimal thermal damage through femtosecond to picosecond laser pulses. This study explores the autonomous optimization of the critical process parameters pulse energy and pulse overlap. Traditional methods for characterizing materials are often labor-intensive and time-consuming. In contrast, we propose a fully automated Bayesian optimization procedure to streamline material characterization, focusing on maximizing specific removal rates while minimizing surface roughness. An experimental setup at the Chair for Laser Technology at RWTH Aachen integrates advanced sensors and a microservice-based software platform to facilitate real-time optimization. The results demonstrate the efficiency of Bayesian optimization in exploring multi-objective parameter spaces, generating a Pareto front that balances productivity and quality. Limitations of the approach are discussed as well.