ADAPTIVE ROBOTIC CARVING - INNOCHAIN
The research, named Adaptive Robotic Carving, challenges the linear progression from the design intention to its materialisation and focuses on the integration of timber material performances within design to manufacturing workflows. It examines the training of a robotic fabrication system based on a combination of sensor feedback collection and different machine learning strategies.
Crafts practices strongly depend on knowledge and material understanding acquired through years of experience. The project seeks to establish a dialogue between different creative human experts and emergent technologies involved in the process, where the instrumental and material knowledge of skilled human craftsmen is captured, transferred, robotically augmented and finally integrated into an interface that makes this knowledge available to designers.
This enables the exploration of novel design opportunities based on an extended range of newly available manufacturing processes and tools which has been trained to adapt to material properties such as the density and grain structure of specific wood species or local features such as knots.
The core proposition is using real-world fabrication data, collected both by human experts and autonomous robotic sessions, to achieve a more accurate geometrical prediction of tools operations on a heterogeneous material such as timber rather than relying on conventional digital simulation methods unable to deal with non-standard fabrication tasks.
One of the outputs of the research, in collaboration with ROK Architects, is the furniture piece Kizamu, Japanese for “carving”, which is composed of a series of carved wooden platforms to display a collection of art pieces in a gallery. Each carved board, while following a similar design logic to the others, presents local individual features and changes in the pattern arrangements due to the control of input design parameters and wood grain behaviours. The trained system has been used as a negotiation platform between top-down design requirements and bottom-up material features, supporting decision-making procedures based on an accurate simulation interface.
Brugnaro, G., Figliola, A., Dubor, A. 2019. Negotiated Materialization: Design Approaches Integrating Wood Heterogeneity Through Advanced Robotic Fabrication. In: F. Bianconi and M. Filippucci, eds. Digital Wood Design: Innovative Techniques of Representation in Architectural Design. Springer International Publishing.
Brugnaro, G., Hanna, S. 2018. Adaptive Robotic Carving: Training Methods for the Integration of Material Performances in Timber Manufacturing. In: J. Willmann, P. Block, M. Hutter, K. Byrne, T. Schork, eds. Robotic Fabrication in Architecture, Art and Design 2018. Zurich, CH: Springer, pp. 336-348.
Brugnaro, G., Hanna, S. 2017. Adaptive Robotic Training Methods for Subtractive Manufacturing. In: T. Nagakura, S. Tibbits, C. Mueller, eds. ACADIA 2017: Disciplines and Disruption, Proceedings of the 37th Annual Conference of the Association for Computer Aided Design in Architecture (ACADIA), Cambridge, MA: Acadia Publishing Company, pp. 164–169.
The project is part of a Ph.D. research conducted by Giulio Brugnaro, supervised by Prof. 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 program under the Marie Sklodowska-Curie grant agreement No. 642877. The project “Kizamu” has been completed in collaboration with ROK Architects (Supervisor: Silvan Oesterle) as part of an industry secondment organised within such framework.