Our Research
Research in the DIGIT Lab is driven by one overarching question: How can AI augment our ability to design better materials, structures, and engineered systems? Our work focuses on two sides of AI-driven design.
First, we use AI to push the limits of what can be designed, centered on two challenges:
Complexity
Advances in modern manufacturing have unlocked unprecedented design freedom, but how can we effectively explore the resulting massive design space?
Creativity
Human creativity is shaped and often limited by individual knowledge, experience, and cognitive biases, which makes innovation slow and unpredictable. Can modern AI help us discover high-performing, "outside-the-box" solutions that people might never consider?
Second, we want to keep humans at the center of the design process by addressing:
Trust
AI can generate design solutions, but when and why should we trust them?
Knowledge
Beyond the solutions themselves, what can designers learn from AI (new design principles, insights, and intuition)?
We pursue these questions by developing computational methods for generative design and design under uncertainty, drawing on machine learning (especially generative modeling), probabilistic modeling and uncertainty quantification, and optimization. Often in collaboration with experts in mechanics, materials, chemistry, and manufacturing, we apply these methods to problems ranging from aerodynamic shapes and active structures to engineered materials (e.g., metamaterials, polymers, and 2D materials).
For more information, you can explore our paper collection or check out our open-source code.