Autodesk Research
Project Dreamcatcher

Project Dreamcatcher

What if a CAD system could automatically generate tens, hundreds, or even thousands of design options that all meet your specified high-level goals? It’s no longer what if: it’s Project Dreamcatcher, and it’s the next generation of computational design.

Dreamcatcher is a goal-directed design (GDD) system that enables designers to input specific design objectives, including functional requirements, material type, manufacturability, performance criteria, and cost restrictions. The infinite computing power of the cloud then takes over.

Dreamcatcher interprets design intent from the objectives specified by the designer and uses the cloud to create thousands of valid design options that meet the designer’s criteria, recommending the best-performing versions for further consideration. It’s not so much about developing a solution as it is about searching for and finding one. This is 180 degrees from the traditional CAD/SIM/optimize workflow.

But don’t worry, GDD doesn’t replace the designer—far from it. It does the grunt work, processing and evaluating design tradeoffs at a speed impossible for humans. What it does do, however, is free up the designer to innovate and create—to move away from repetitive design tasks and calculations and instead focus on creative design.

This is cloud computing in its purest form; true computing rather than simple file storage. Previously only available to institutional and government-owned agencies with supercomputers, it’s now available to anyone.

Moving beyond the traditional design workflow that begins with geometry-based input, GDD instead takes the design problem as input, using evolutionary algorithms to generate solutions by mimicking the way nature accepts or rejects designs. This is particularly useful in manufacturing. Traditional software tools have been unable to adequately envision optimal solutions to the increasingly complex and highly dimensional problems found in today’s manufacturing space. Now, objects can be designed that optimally suit their function; objects that were once impossible to manufacture.

It’s a new workflow: a fusion of the designer, artificial intelligence, and the cloud. It’s Project Dreamcatcher—design’s next great leap.

Project Dreamcatcher

Project Dreamcatcher

Recent Publications

Natural Language Problem Definition for Computer-Aided Mechanical Design

Hyunmin Cheong, Wei Li, Li Shu, Alex Tessier, Erin Bradner & Francesco Iorio. (2014).
Natural Language Problem Definition for Computer-Aided Mechanical Design
CHI-DSLI Workshop 2014 Conference Proceedings:
DSLI Workshop at ACM CHI.
4 pages.

Download PDF | Details

Parameters Tell the Design Story: Ideation and  Abstraction in Design Optimization

Erin Bradner, Francesco Iorio & Mark Davis. (2014).
Parameters Tell the Design Story: Ideation and Abstraction in Design Optimization
SimAUD 2014 Conference Proceedings:
Symposium on Simulation for Architecture and Urban Design.
8 pages.

Download PDF | Details

All Text and Imagery Copyright © 2014 Autodesk, Inc. All Rights Reserved.