We approach Mixed Reality from a Human-Computer Interaction (HCI) perspective. We don’t just build applications; we engineer the underlying methods that make them effective, accessible, and measurable. A primary pillar of our work is the development of Adaptive Training Systems and intelligent agents that adapt to the user. For instance, in ASTRA, we investigate anthropomorphic AI agents for soft skill development, while TwinMaP utilizes digital twins to create adaptive work instructions that evolve based on employee feedback. Similarly, Sidekick focuses on accessible digital assistance for sheltered workplaces.

A core mission of our group is removing “programming barriers” through authoring tools for immersive media. We build open-source authoring tools that empower domain experts to create their own immersive content without writing code. This includes visual scripting for handheld AR in TrainAR, intuitive VR editors for nursing education in ViRDiPA, and mobile training scenarios for midwifery in Heb@AR. By integrating eye-tracking and physiological sensors in projects like ReqET, and developing dedicated analysis platforms like SUS.tools, we try to scientifically quantify “cognitive load” and usability and validate the effectiveness of our training scenarios across the board.