KO

Self-Driving Laboratories

Investigating self-driving workflows that connect laboratory automation with active learning.

Bayesian optimization concept linking experimental results, predictive models and subsequent experiments
Bayesian optimization concept linking experimental results, predictive models and subsequent experimentsView original image ↗

We investigate self-driving workflows that connect experimental design, execution, analysis and selection of the next conditions. Bayesian optimization links experimental results to predictive model updates and recommendations for subsequent experiments. Our research explores data flows and decision frameworks for integrating this cycle with laboratory automation.

WHY IT MATTERS

Structuring the research workflow allows each experiment to inform the next decision.

Research directions

  • Laboratory automation and analytical integration
  • Active learning and Bayesian optimization
  • Building and validating closed-loop experiments
← All core technologies