Industrial knitting machines are programmable systems that can fabricate complex soft objects. However, this traditional manufacturing system has largely been overlooked as a technology for custom and rapid fabrication. For machine knitting to be widely adopted for custom fabrication, end users must be allowed to work in an intuitive design space instead of low-level machine operations.

In this thesis, I lay out foundational tools for machine knitting that allow users to think about what they want to create in terms of 3D shapes instead of how the machine constructs by showing that programming for machine knitting can be organized to decouple high-level design challenges from low-level machine input decisions.

Such an organization allows exploration of various aspects of machine-knitting independently: pattern design, layout planning, and machine-code generation. I classify the space of shapes that can be constructed with industrial knitting machines. I present a new data-structure to represent 3D shapes as knitting programs. I describe geometric algorithms to create patterns from 3D models and an editing framework to modify patterns in 3D. For fabrication, I describe a scheduling algorithm to translate patterns into low-level machine code. Together, these tools and techniques allow designers and end-users to treat 3D machine knitting as an accessible 3D-printing-like soft fabrication system.

122 pages

Thesis Committee:
James McCann (Chair)
Jessica Hodgins
Keenan Crane
Adriana Schulz (University of Washington)

Srinivasan Seshan, Head, Computer Science Department
Martial Hebert, Dean, School of Computer Science

Return to: SCS Technical Report Collection
School of Computer Science

This page maintained by [email protected]