Knitting is the intermeshing of yarns into loops that leads to the creation of fabrics. Fabrics knitted with several different yarns constitute a modern class of advanced engineering materials. Additionally, computer-controlled knitting offers great promise as a digital manufacturing technology because of the ability to make novel materials from multiple yarn types and varying loop patterns. Given the enormous flexibility of modern knitting equipment and the variety of yarn materials currently available, knitted structures can now be viewed as a versatile, programmable, multi-scale and microstructure-based engineering material system that can be designed and manufactured to meet a wide range of functional and mechanical requirements. While the current state of the art in commercial knitting technology is quite capable of designing and manufacturing a variety of materials for consumer clothing, many significant technological gaps exist that prevent knitting from becoming a general process for digital manufacturing. The critical barrier to the production of these versatile material systems is the lack of validated comprehensive design tools and accurate predictive modeling methods. This project will develop an integrated design framework that addresses these barriers in order to enable the programmable manufacturing of customized knitted materials. Developing advanced technologies in this field has the potential to create new, revitalized economic activity in an American industrial sector that has mostly known decline in recent decades.
The project has several technical objectives. They are: a) the creation of a design framework that determines yarn specifications and processing commands given the performance requirements for a knitted material, b) the development of a data-driven algorithm for computing inverse mappings from low-dimensional requirements space into high-dimensional control data space that utilizes efficient nearest-neighbors search and scattered data interpolation techniques to find viable design solutions for knitted materials, c) the definition of an objective function capable of determining if knitting design requirements have been fulfilled, d) the utilization of information pathways that allow for an inverse-forward simulate-manufacture-evaluate-validate feedback loop in the design process, and e) the integration of design/simulation processes with the production, testing and evaluation of actual materials to provide an accelerated, programmable manufacturing capability for novel, customized knitted materials.
