Autodesk introduces Design Graph, a machine learning system integrated in Autodesk A360 that uses algorithms to extract large amounts of rich 3D design data from industrial design and mechanical engineering files. Design Graphs gives computers the ability not just to think, but to learn independently, identify and understand designs based on their inherent characteristics, their shape and structure, rather than by labeling, tags or metadata given by users, that tends to be unreliable.
Using fast and simple Google search-like functionality for the world of 3D models, the system allows to search pre-existing designs from a firm’s design files database, selected according to required characteristics such as name, shape, category, properties or a combination of these, then ready to be dropped into a new design. The solution is an initial step for Autodesk in harnessing the powerful technology of machine learning, to improve and accelerate the design process, enabling designers to focus more on solving problems rather than the mechanics of representing their design, saving valuable time, eliminating redundant work and reducing costly errors.