Augmented Drawn Construction Symbols: A Method for Ad Hoc Robotic Fabrication

Jens Pedersen, Narendrakrishnan Neythalath, Jay Hesslink, Asbjørn Søndergaard, Dagmar Reinhardt

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

Abstract

The global construction industry is one the least productive over a 30-year period [1], which arguably could be related to virtually no implementation of digital and automation technologies within the construction industry [2]. Construction processes arguably consist of expensive manual labor or manual operation of mechanized processes, where hand-drawn markings on work-objects or partly build structures are used to inform and steer the construction process or allows for ad-hoc adjustments of elements. As such, the use of on-object, hand-drawn information is considered integral to the modus operandi of a plurality of construction trades, where timber construction and carpentry are of special interest. In contrast, emerging methods of digital production in timber construction, implicitly or explicitly seek to eliminate the interpretive component to the construction work, imposing a top-down paradigm of file-to-factory execution. While such systems offer a performance, increase compared to manual labor, it is notoriously sensitive to construction tolerances and requires a high level of specialism to be operated, which could alienate craft educated workers.
This research argues that developing methods for digital production compatible with on-site human interpretation and adaptation can help overcome these challenges. Additionally, these methods offer the opportunity to increase the robustness and versatility of digital fabrication in the context of the construction site.
The paper reports on a new method titled Augmented Drawn Construction Symbols (ADCS) that through a visual communication system converts on-object hand-drawn markings (OOHDM) to CAD drawings and sends them to a robotic system. The process is demonstrated on a full-scale prototypical robot setup.
OriginalsprogEngelsk
TidsskriftInternational Journal of Architectural Computing
Vol/bind18
Udgave nummer3
ISSN1478-0771
StatusUdgivet - sep. 2020

Kunstnerisk udviklingsvirksomhed (KUV)

  • Nej

Citationsformater