3D Robot Knit Lab

The work stems from an attempt to translate a mathematical construct into a knitted artifact through controlled machine processes. The starting point is the Hilbert space-filling curve, selected for its ability to map a one-dimensional sequence onto a two-dimensional plane while preserving locality. This mapping is operationalized by encoding directional states (straight, left, right) as discrete color and stitch “packages,” which are then sequenced into a compressed knitting pattern. The project is therefore driven by a constructive logic: geometry is not represented visually but executed materially through knitting instructions. The relationship to mathematics is structural rather than referential. The Hilbert curve functions as a generative system that defines both form and process. Its iterative nature is translated into repeatable knitting units, while the curve’s compression into a linear sequence aligns with machine-readable patterning. A later reflection identifies this sequence as equivalent to an L-system representation, indicating that the garment operates as a physical instantiation of a formal grammar. Additionally, quantitative control is maintained through parameters such as stitch count, loop length, and yarn consumption, linking discrete mathematics to textile fabrication. The uniqueness of the fashion item lies in coupling algorithmic structure with material manipulation across the value chain. The scarf is shaped not only by short-row knitting to produce curvature but also by a custom dye process that inverts conventional space dyeing. A full-length knitted tube is dyed as a single gradient and then unraveled, producing a non-repeating color sequence aligned with the fractal path. This ensures that chromatic variation follows the encoded geometry rather than a periodic pattern. The result is a textile in which form, color, and production logic are co-dependent, with the machine acting as an interpreter of a mathematical system.