In mid-March, Pro²Future was represented at this year’s HRI 2026, International Conference on Human-Robot Interaction.
As a central platform for Human Robot Interaction, HRI 2026 brought together international experts to discuss the future of human-centered AI and Human-Robot Interaction.
Explainability as a Key Enabler of Human-Robot collaboration
At the conference, Verena Szojak presented her latest research on explainability needs in human-robot collaboration, addressing one of the most pressing challenges in industrial AI.
The study investigates real-world interaction in a bicycle assembly task using a 6-axis collaborative robot (cobot). By applying think-aloud protocols and semi-structured interviews, the research provides detailed insights into how humans perceive and interact with cobots in dynamic environments.
Key Findings
The results highlight three central requirements for effective collaboration between humans and robots:
- Continuous feedback on the cobot’s current state is essential for situational awareness
- Context-sensitive guidance is preferred over complex or generic system explanations
- Simple and intuitive communication formats, such as short text cues and light signals, significantly improve usability
From Explainability to Practical Implementation
Explainability in human-robot interaction is not about increasing the amount of information provided. Instead, it is about delivering the right information, at the right time, in the right format.
This perspective is critical for transferring research into industrial practice, where usability, trust and efficiency directly impact system adoption.
Contribution to Cognitive and Sustainable Production Systems
The presented work directly contributes to the mission of Pro²Future:
Designing cognitive and sustainable products and production systems that empower humans in real industrial environments.
By focusing on explainable and human-centered AI, Pro²Future supports the development of next-generation production systems that are not only technologically advanced, but also intuitive, trustworthy and deployable at scale.
Further Information
Read the full publication:
https://dl.acm.org/doi/10.1145/3776734.3794435