Adaptive robotic carving is an innovative technique that uses artificial intelligence and sensors to predict and simulate the outcome of the interaction between non-standard tools and heterogeneous materials, such as wood. This technique allows designers to explore a wide range of creative and sustainable solutions, informed by the properties and behaviours of materials and tools.
The project is part of a doctoral research carried out by Giulio Brugnaro, supervised by Professor Bob Sheil and Dr Sean Hanna, at the Bartlett School of Architecture, University College of London, within the framework of the “InnoChain Training Network”, supported by the European Union’s Horizon 2020 research and innovation programme.
The aim of this technique is to integrate manufacturing knowledge early in the design process by collecting real-world manufacturing data, using different sensing devices and machine learning models to achieve accurate prediction of carved geometries, informed by material behaviours. In this way, feedback can be established between different stages of the process, allowing designers to explore a wider variety of manufacturing methods and materials, often leading to more efficient and less wasteful solutions.
The acquisition of manufacturing data was structured through a series of recording sessions to store, in a library of datasets, the combination of manufacturing parameters and their respective results generated by different material properties (e.g. grain structure, density, direction), wood species and carving tools. The collected datasets can be used to train multiple artificial neural networks (ANNs), whose main objective is to predict the carved geometry generated by a user-defined robotic trajectory and a set of manufacturing parameters.
The curation of the training process, by a team of designers, represents the cornerstone of the design workflow, as the selection of relevant material properties and manufacturing parameters directly determines the range of solutions available later in the digital design exploration. In this way, the trained system represents a knowledge package that can be integrated within an interface to digitally evaluate multiple design solutions, which would not otherwise be available, informed by the properties and possibilities of tools and materials.
The adaptive robotic carving technique has been applied to a number of collaborative projects with industrial partners, such as ROK Architects, Zurich, and BIG, Copenhagen, to develop a catalogue of design explorations for a wide range of applications, from furniture to building components of larger assemblies. These projects demonstrate the creative and innovative potential of adaptive robotic carving as a sustainable and efficient form of digital manufacturing.
The adaptive robotic carving project uses robotic arms from the company KUKA, which are market leaders in automation and robotics. Specifically, the KR 120 R2500 pro and KR 210 R3100 ultra models are used, which have a reach of 2,500 and 3,100 mm respectively, and a load capacity of 120 and 210 kg. These robotic arms are equipped with non-standard carving tools, such as knives, saws or planers, which are adapted to the properties and behaviour of the wood.
If you require more information, please don’t hesitate to contact us: we will be ready to help you.