Our group studies a range of unsteady flow phenomena leveraging data science, nonlinear machine learning, complex network theory, information theory, and computational fluid dynamics. Our ultimate goal is to build a data-oriented foundation for real-time analysis, modeling, and control of unsteady flows ubiquitously appearing in various situations around small air vehicles, airplanes, motor vehicles, and fluid-based industrial machines.

Equipped with nonlinear machine learning-based sparse sensor reconstruction and data compression supported through traditional numerical and experimental analysis, our approach enables high-resolution reconstruction, real-time prediction, and control of flow fields with limited availability of data.
These techniques are aimed at analyzing and controlling large-scale, complex nonlinear flow phenomena that have been challenging to tackle with conventional linear methods.

・Real-time spatiotemporal flow field reconstruction from sparse sensors is enabled by turbulence super-resolution analysis with machine learning.
・Understanding and modeling of unsteady fluid flows at low cost is made possible through low-dimensional manifold identification and compression.
・Development of explainable machine-learning approaches for analyzing causal vortex interactions based on complex network theory and information theory.
・Multi-modal data analysis through the fusion of numerical, experimental, and theoretical data.

Our group aims to develop technologies that accurately sense, predict, model, and control fluid flows —such as air and water— around objects including airplanes, automobiles, and wind turbines, even with sparse sensor information.

These technologies can contribute to society in various ways, including:
・Improving fuel efficiency and safety of aircraft
・Enhancing the aerodynamic performance of vehicles for energy savings
・Supporting disaster prevention through wind flow prediction during emergencies

We actively seek to co-create innovations through joint research with industrial companies interested in the following areas:

・Predicting and controlling fluid flows using AI and machine learning
・Understanding flow structures through information theory and network science
・Building highly accurate and reproducible models by integrating traditional fluid dynamics with modern data-driven methods

Equipped with physics-based nonlinear machine learning, we are working to develop groundbreaking fluid analysis technologies that benefit a wide range of industrial, environmental, and societal applications.

• Data Science
• Artificial Intelligence
• Fluid Dynamics
• Turbulence
• Aerodynamics
• Machine Learning
• Information Theory
• Network Theory
• Fluid Engineering
• Aerospace Engineering
• Data Fusion
• Digital Twin
• Reduced-order modeling