Hosmer, Tyson;
Wang, J;
Jiang, Wanzhu;
He, Ziming;
(2022)
Integrated Reconfigurable Autonomous Architecture System.
In:
Proceedings of the 42nd Annual Conference of the Association for Computer Aided Design in Architecture: Distributed Proximities, ACADIA 2022.
(pp. pp. 1-14).
Acadiahttps://2022.acadia.org/
Preview |
Text
IRAAS_Hosmer_Accepted.pdf - Accepted Version Download (13MB) | Preview |
Abstract
Advances in state-of-the-art architectural robotics and artificially intelligent design algorithms have the potential not only to transform how we design and build architecture, but to fundamentally change our relationship to the built environment. This system is situated within a larger body of research related to embedding autonomous agency directly into the built environment through the linkage of AI, computation, and robotics. It challenges the traditional separation between digital design and physical construction through the development of an autonomous architecture with an adaptive lifecycle. Integrated Reconfigurable Autonomous Architecture System (IRAAS) is composed of three components: 1) an interactive platform for user and environmental data input, 2) an agent-based generative space planning algorithm with deep reinforcement learning for continuous spatial adaptation, 3) a distributed robotic material system with bi-directional cyber-physical control protocols for simultaneous state alignment. The generative algorithm is a multi-agent system trained using deep reinforcement learning to learn adaptive policies for adjusting the scales, shapes, and relational organization of spatial volumes by processing changes in the environment and user requirements. The robotic material system was designed with a symbiotic relationship between active and passive modular components. Distributed robots slide their bodies on tracks built into passive blocks that enable their locomotion while utilizing a locking and unlocking system to reconfigure the assemblages they move across. The three subsystems have been developed in relation to each other to consider both the constraints of the AI-driven design algorithm and the robotic material system, enabling intelligent spatial adaptation with a continuous feedback chain.
| Type: | Proceedings paper |
|---|---|
| Title: | Integrated Reconfigurable Autonomous Architecture System |
| Event: | Annual Conference of the Association for Computer Aided Design in Architecture (ACADIA) |
| Location: | Philadelphia, PA, USA |
| Dates: | 27th-29th October 2022 |
| Open access status: | An open access version is available from UCL Discovery |
| Publisher version: | https://2022.acadia.org/ |
| Language: | English |
| Additional information: | This version is the author accepted manuscript. For information on re-use, please refer to the publisher's terms and conditions. |
| UCL classification: | UCL UCL > Provost and Vice Provost Offices > UCL BEAMS UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of the Built Environment UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of the Built Environment > The Bartlett School of Architecture |
| URI: | https://discovery-pp.ucl.ac.uk/id/eprint/10165846 |
Archive Staff Only
 |
View Item |
