September, 2025

We Built a Robot Librarian - and Won in Luxembourg

Last week, our team traveled to Luxembourg for the BNL Project demo day, an event that brought together innovators working on an ambitious idea: a robot librarian. The concept sounds simple at first - a robot that moves through a library, scans books on shelves, and checks whether each book is in the right place. But in practice, it’s an incredibly complex challenge, combining robotics, computer vision, and algorithms that need to work in real-time.

From Setbacks to Success

Our journey wasn’t straightforward. On the first demo day, we faced unexpected technical issues. But instead of stopping there, we regrouped, fixed critical bugs, and pushed harder to improve the system. By the time we returned for the second and final demo day, our solution had transformed.

What We Built

To make the robot librarian work, we combined:

• Computer vision techniques that allow the system to recognize books on shelves, even when placement is imperfect.
• A shelf verification process that checks whether books are in the right order.
• System optimizations that improved both speed and reliability.

By demo day, the system reached a high level of accuracy in detecting and validating book placement - strong enough to demonstrate the feasibility of the concept.

The Result

When the final evaluations came in, our team was thrilled - we had won in the demo day competition 🎉.
📢 You can see the official announcement here: GovTech Lab Luxembourg on LinkedIn.

What’s Next

Winning the demo day is just the beginning. The next big milestone is building a pilot version of the robot librarian, taking the concept beyond demonstrations and into a real-world library environment. This will be a chance to see how our robot performs outside controlled tests, and how it can truly support library operations in practice.

To learn more about the project and its vision, you can read the full announcement from the Ministry of Digitalisation: BOT4LIB Proof of Concept.

Final Thoughts

The BNL Project has been more than just a competition - it’s been a glimpse into how robotics and AI can support knowledge, culture, and learning spaces. For us, it was also a reminder that innovation rarely happens in a straight line. Sometimes, the path includes failures, retries, and unexpected breakthroughs - and that’s exactly what makes the success feel so rewarding.

Note: This post shares only publicly available information and general insights. It does not disclose any confidential or proprietary details and remains fully compliant with the NDA agreement.