EAGER/Cybermanufacturing: WYSIWYG (What You See Is What You Get) Middleware for Additive Manufacturing

Massachusetts Institute Of Technology, Cambridge MA

Investigators

Abstract

Custom products can offer advantages over their mass-produced counterparts, including greater comfort, unique aesthetic appeal, and/or better performance. Additive manufacturing promises cost-effective fabrication of such custom products, and there has been steady progress in the additive manufacturing hardware, the range of available materials, and the production cost. However, taking advantage of additive manufacturing for customization requires new types of digital models and software applications. Current solutions to this problem either provide very restricted customization, or do not guarantee that the customized objects are valid and can be fabricated to the desired specifications. This EArly-concept Grant for Exploratory Research (EAGER) project award supports research to establish a software framework for additive manufacturing that provides a substantially easier and a more scalable way of developing domain-specific software applications supporting an interactive, highly-predictive preview of digital objects. The design exploration tools and accurate preview mechanisms will enable additive manufacturing according to precise dimensional specifications. This translates to enormous time and material cost savings, accelerating the adoption of additive manufacturing. This software framework will be made available to a large population of expert designers, engineers, hobbyists, and students, facilitating design exploration, manufacturing, and customization of both industrial and consumer products. The software framework will be a middleware that allows development of domain-specific applications for digitally-driven manufacturing, and will be informed by quantitative knowledge of the design space that can be realized by a given additive process. To realize this goal, the following components are to be designed and built: (1) general representations for physically realizable, end-user customizable digital object families encapsulated within domain-specific applications; (2) data-driven methods for an accurate and computationally efficient characterization of additive manufacturing processes, including identification and mitigation of geometric imperfections, feature size bounds, and intrinsic process defects; (3) methods for an interactive and highly-predictive exploration of the design space spanned by a given object family and a desired manufacturing process; and (4) computationally tractable methods for optimizing process parameters for a desired digital object and a specific additive manufacturing material-machine combination.