Human-In-The-Loop

Collaboration between humans and artificial intelligence is at the centre of our research on human-in-the-loop. We develop methods and technologies that enable safe, trustworthy and efficient interaction between human experts and AI systems throughout the entire life cycle. This is particularly important in safety-critical applications in robotics, transport and aerospace.

To optimise this collaboration, our interdisciplinary research team is developing innovative solutions that combine the precision and speed of AI systems with the intuition, experience, and judgement of human specialists. Our research focuses on how humans can influence the learning process of AI systems, and on how to understand, evaluate and correct AI decisions without compromising the system's overall efficiency and reliability.

Based on these findings, we develop adaptive interfaces and interaction concepts for our projects. These enable continuous monitoring of AI systems, targeted influence over their learning processes, and intervention if necessary. Particular focus is placed on ensuring the safe interaction between humans and AI, with the aim of guaranteeing reliable functionality even in unforeseen situations or in the event of failures.

This systematic approach forms the basis for trustworthy AI applications by integrating human-in-the-loop methods into AI development. These fulfil the high safety standards in robotics, transport and aerospace. At the same time, they offer the necessary adaptability to handle complex and changing application scenarios.