House_n > Projects >
This project investigates how machine learning techniques can be used to create computerized architectural design critics. Web based design critics can automatically provide non-experts with feedback based upon a particular architect's style. This system has been prototyped using the Platform for Consumer Driven Participative Design of Open (Source) Buildings described above.
Advances in building and computational technologies, coupled with a reorganized and integrated system of residential design, may make personal environments tailored to the needs of the individual a possibility for a larger segment of the population. This project explores the essential first step towards ubiquitous personalized design: the development of tools to help non-expert designers identify their perspective, needs, and goals. The resulting information can then be linked to new design algorithms and a more rational and integrated just-in-time manufacturing process.
The construction of a new home in the US typically consists of 80% field labor and 20% material costs. Reversing this ratio will allow four times as much money to be devoted to materials, design, engineering, safety, and technologies in the home. Borrowing from recent innovations in the automobile industry, researchers are developing concepts for creating buildings from an integrated "chassis" that can be rapidly and precisely installed with minimal field labor. In one integrated assembly, pultrusion glass fiber composite beams and columns provide structure, insulation, sensor arrays, lighting, signal and power cable raceways, and ductwork. The chassis provides the necessary physical, power, and signal connections for mass customized infill components to be quickly installed, replaced and upgraded without disruption. An alternative steel-frame and concrete volumetric chassis is currently under development for mid-rise multifamily buildings.
House_n Research Group