Artifex Automata

The "Artifex Automata" project explores the intersection of traditional craftsmanship and advanced digital technology to address the challenges of conserving historic buildings. As skilled craftspeople become increasingly scarce, this research aims to utilize robotics and machine learning to preserve and improve the craft of stonemasonry, which is critical to the restoration and adaptive reuse of heritage buildings. By integrating a machine learning model trained on the experience stonemasons into a robotic arm, the project demonstrates that this system is capable of predicting and executing precise chiseling work on sandstone that is very similar to the work of a human craftsperson.

The research focuses on developing a workflow that enables the robotic system to replicate the intricate and nuanced chiseling techniques traditionally performed by stonemasons. This is achieved through a combination of motion capture, 3D scanning and machine learning, whereby the robot learns to perform the chiseling based on data collected from a trained stonemason. The results show that while the robot can produce surfaces similar to those made by hand, there are subtle differences that challenge previous notions of craftsmanship and robotic precision. These results open up new possibilities for the integration of digital technologies in heritage conservation and represent a sustainable method of preserving historic buildings while respecting the craftsmanship that defines them.

This project, carried out as a student project in the "Computation in Architecture" master program at CITA, Royal Danish Academy, not only demonstrates the technical feasibility of using robots in stonemasonry, but also promotes a paradigm shift in the way craft skills can be perceived, preserved and passed on. By bridging the gap between traditional craftsmanship and modern technology, the project advocates a future in which digital tools and techniques support and preserve the rich heritage of manual crafts.