Steven Brunton
University of Washington, WA·Engineering
Aircraft and advanced manufacturing systems are among the most complex machines people build, yet designing, testing and producing them is still slow and expensive. These systems obey the laws of physics, such as fluid flow and structural mechanics.
- Fields
Engineering
Computer Science
Mathematics & Statistics
- Topics
physics-informed machine learning
physics-aware AI
artificial intelligence
- Grants
- 1 known grant, through Sep 2031
What they work on
- Develop physics-aware machine learning methods that build physical laws and stability constraints into the models, producing interpretable models that work reliably for aerospace design and control.
- Build uncertainty-aware digital twins, continuously updated virtual copies of engineered systems, that combine simulation, sensing and data across design, testing and manufacturing to support real-time decisions and adaptive control.
- Create AI copilots for engineering design, including domain-specific large language models that help with automated reasoning, hypothesis generation, model validation and data consistency checks.
- Construct benchmark infrastructure for evaluating AI systems on dynamic physical problems, including turbulence and aerodynamic shape optimization.
More about the research
That raises a question: can artificial intelligence (AI) learn those laws well enough to help engineers predict, optimize and control such systems reliably? This multi-thrust AI institute is built around that question. It joins AI, engineering, mathematics and the physical sciences, and it tests the resulting tools on real aerospace problems with industry and national laboratory partners.
Why it matters
Faster, more reliable design and testing could cut the time and cost of bringing new aircraft and other aerospace technologies from concept to use, while improving safety, performance and efficiency. Aerospace manufacturers, commercial aviation and defense would be the main beneficiaries. Open software, public datasets and benchmarks could also help other scientists and engineers judge whether AI tools are trustworthy for physical systems.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Michael Alvarez
University of Puerto Rico-Rio Piedras, PR·Biomedical Engineering
Develop a scalable, AI-enabled bioreactor platform for growing mesenchymal stem cells (adult stem cells used in therapies) and collecting their extracellular vesicles, small particles cells release.
Mitsunori Ogihara
University of Miami, FL·Computer Science
Develop accelerated methods for analyzing medical images, using AI/ML together with coupled software and programmable hardware so results arrive fast enough for real-time use.
Qi Yu
University of California-San Diego, CA·Computer Science·Early-career
Develop deep learning models that forecast spatiotemporal dynamics while obeying physical laws, with the goal of simulating ocean currents, traffic and epidemic spread faster than numerical solvers.
The school
University of Washington: 81 labs on PI Pages, $1.7B a year on research, 694 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.