Data-Driven Dynamic Systems and Controls for Engineering
NEXT START DATE: Fall 2025
PRIORITY DEADLINE: May 1, 2025
APPLY BY: July 1, 2025
Application opening soon
LOCATION
Online
DURATION
15 months part-time
TIMES
Mostly asynchronous
TOTAL COST
$18,000 (estimated)
Program overview
The Graduate Certificate in Data-Driven Dynamic Systems and Controls for Engineering offers engineering professionals the opportunity to transform their careers with the skills to model and control dynamic systems using a data-centric approach. Students can choose to take the certificate independently or combine it with another eligible data-intensive certificate in the College of Engineering to obtain a flexible Master of Science in Artificial Intelligence and Machine Learning for Engineering.
Who is this program for?
This certificate is for engineers who want to leverage data to improve modeling and control of dynamic systems in domains such as manufacturing, robotics, fluids, and materials. Students will advance their careers by building on their traditional engineering expertise and learn to use machine learning and AI to perform system identification, optimize sensor placements, and responsibly automate engineered systems.
Learning outcomes
-
Implement and evaluate data-driven methods
Choose and implement data-driven methods that are appropriate for dynamic systems modeling and automatic control, considering the challenges of safety-critical and mission-critical systems.
-
Build foundational machine learning skills
Strengthen math and coding skills, creating a foundation that enables you to stay up to date with new machine learning and data-driven techniques throughout your career.
-
Communicate methods and results
Practice and receive feedback on communicating your work to a variety of people through data visualization, verbal presentations, and written reports.
Courses
The Graduate Certificate in Data-Driven Dynamic Systems and Controls for Engineering is an 18-credit certificate consisting of the following: 5-credit required foundations course, 4-credit core course, 5-credit core course which includes an applied project, 4 credits of electives.
Foundational skills in math, dynamics, controls, and coding. Includes differential equations, linear control, optimization, and best practices for scientific software development.
Techniques for modeling time-dependent systems and discovering interpretable dynamic models using machine learning.
Optimization of control laws using machine learning and techniques for modeling controlled and forced dynamic systems using data.
Reconstruction of signals from sparse measurements, optimization of sensor choice and placement, active learning, and advanced filtering and denoising techniques.
Certificate stackability
Admissions and application
Applicants will need to have a 3.0 cumulative grade point average on a 4-point scale from an accredited school and have met specific coursework requirements.
Meet your instructors
Steve Brunton
Professor, Department of Mechanical Engineering & Adjunct Professor, Applied Mathematics
Brunton serves as the Director of the AI Center for Dynamics & Control and holds the position of Data Science Fellow at the eScience Institute. His research focuses on combining techniques in dimensionality reduction, sparse sensing, and machine learning for the data-driven discovery and control of complex dynamical systems. Additionally, he develops adaptive controllers using machine learning within an equation-free framework. His work spans applications in fluid dynamics, such as closed-loop turbulence control, as well as in neuroscience, medical data analysis, networked dynamical systems, and optical systems.
Krithika Manohar
Assistant Professor, Department of Mechanical Engineering
Manohar’s research focuses on developing algorithms for data-driven prediction and control of complex dynamical systems. Her work uses dimensionality reduction techniques rooted in operator theory and manifold learning to discover physically meaningful features from data, and optimize sensors and actuators for downstream decision-making (sparse sensing). Target applications of sparse sensing optimization include fluid flow reconstruction, control, image recovery, and aircraft manufacturing.
Michelle Hickner
Curriculum and Instruction Lead, AI Education & Research Initiative
Hickner has served in a variety of teaching roles at the University of Washington, focusing on hands-on and experiential learning. Her scholarly interests include engineering education, sensing and control in animal flight, and data-driven system identification. Her past work has included running the UW mechanical engineering composite shop, and research and development of devices for airborne particle handling.