AI Robot Solves Ancient Puzzle in Pompeii

Related Vendors

Rheinische Friedrich-Wilhelms- Universität Bonn Dez.8 - Hochschulkommunikat.

FAULHABER Drive Systems

Wuxi InfiMotion Technology Co., Ltd.

Archaeologists could only dream of this for a long time: An "intelligent robot" piece by piece reassembles broken ancient wall paintings from Pompeii.

A research team led by Prof. Dr. Maren Bennewitz from the University of Bonn (Germany) taught the robot how to solve puzzles. It is part of the European collaborative project Re-Pair ("Reconstructing the Past: Artificial Intelligence and Robotics Meet Cultural Heritage"), which was recently completed.

Precisely Placed Thanks to AI-Based Planning Algorithms

In the Re-Pair project, the Bonn team focuses on the robotic system. The team developed planning algorithms for the robot's bimanual arm movements, enabling it to place fresco fragments with great precision.

Maren Bennewitz, Professor of Humanoid Robots and Prorector for Digitalization and Information Management: "In Re-Pair, we contribute our many years of experience with motion planning and manipulation by robots. The goal of the project was to investigate whether robots can take over the tedious sorting and puzzle work so that experts can spend their time where human expertise is indispensable."

The University of Bonn (Germany) was also responsible for the integration and evaluation of the entire robotic platform, ensuring that all components fit together seamlessly. To achieve this, doctoral student Nils Dengler spent five weeks in Pompeii.

Robot With Soft Gripping Hands Picks Up Fragments

At the heart of Re-Pair is a robotic platform with two robot arms and soft gripping hands that pick up fragments of wall paintings and place them at the position calculated by the AI puzzle solver. It is precisely at this interface between intelligent planning and physical execution that the Bonn research focuses: "Our algorithms calculate the movements for the two robot arms to reliably pick up the fragments and carefully place them at the calculated position," says Nils Dengler.

AI Solves Fresco Puzzle from Pompeii

The system was tested in the Archaeological Park of Pompeii. Among other things, it holds ceiling paintings from the "House of the Painters at Work" as well as frescoes from the Schola Armaturarum, which were heavily fragmented by the eruption of Mount Vesuvius in 79 AD, as well as bombings during World War II and a collapse in 2010.

The fragments are first digitized using a specially developed 3D scanning system. Based on this, the AI attempts to solve the "puzzle": It suggests suitable combinations between fragments and calculates how larger image segments can be reconstructed from hundreds or thousands of pieces.

The challenge is that, unlike a regular puzzle, we don’t have a picture on the box.

Prof. Bennewitz

"The challenge is that, unlike a regular puzzle, we don’t have a picture on the box," says Maren Bennewitz. "Many pieces are heavily damaged or entirely missing, and fragments from different works are often mixed. This makes it all the more important for robotics and AI to work closely with archaeological expertise."

Automated Reconstruction Relieves Archaeologists

The reconstruction of broken artifacts has so far been a task requiring a lot of patience and often remains unfinished for years in the day-to-day operations of museums. By largely automating the time-consuming steps of digitization, proposal calculation, and physical reconstruction, Re-Pair allows archaeologists to focus more on scientific analysis.

It is fascinating to see how robotics can not only solve tasks in factories or logistics but also help preserve our cultural heritage.

Prof. Bennewitz

About the Project

The European Union-funded project was coordinated by the Ca’ Foscari University of Venice. It combines cutting-edge methods of artificial intelligence and robotics with the needs of archaeology. Other project partners include the Italian Institute of Technology (IIT), the Archaeological Park of Pompeii, Ben-Gurion University of the Negev, and Instituto Superior Técnico Lisbon.

The Re-Pair project was funded under the European Union's Horizon 2020 program, Grant Agreement No. 964854. Further information about the project can be found at www.repairproject.eu

Machine Authentication

Machine Operator Authentication - Why knowing "who" matters