About

We continue to expand the frontiers in aerospace engineering by focusing on Modeling as the keystone in the Multi-Disciplinary Analysis, Optimization, and Control (MDAOC) of advanced transportation systems.

Research topics

• Computer Science
• Mathematics
• Artificial Intelligence
• Human–computer interaction
• Physics
• Programming language
• Mathematical analysis
• Engineering
• Mechanics
• Materials science
• World Wide Web
• Aerospace engineering
• Mathematics education
• Mathematical optimization
• Geometry
• Pedagogy

Selected publications

• Physics-Embedded Data-Driven Modeling for Model Predictive Control of Battery Charging Converter

  IEEE Transactions on Industry Applications · 2026-01-01

  articleSenior author

  Model predictive control (MPC) is widely used for battery converter charging control due to its strong capability to explicitly handle multiple objectives and complex constraints in real time. However, its performance depends on the accuracy of the prediction model, which is often degraded by incomplete prior physical knowledge, parameter uncertainty, and aging effects, thereby necessitating data-driven modeling strategies. Moreover, periodic model updating and real-time control require both low training and inference time costs, which purely data-driven models struggle to achieve. To address these challenges, a physics-embedded dictionary-based system identification (PhD-SI) method based on the sparse identification of nonlinear dynamical systems (SINDy) method is proposed. It integrates physical knowledge with data in a compact nested linear-in-parameter matrix representation equipped with high efficient fractional-power monomial base functions with a closed-form sparse regression solution, enabling accurate system identification with low training and inference time costs under incomplete prior knowledge. Results on a battery charging case study demonstrate improved modeling accuracy, enhanced closed-loop control performance under disturbances compared with neural-network-based methods, and strong robustness to parameter uncertainty, indicating its potential to support battery charging system vendors in developing high-performance charging solutions.

  PublisherDOI

• Data-Enabled Predictive Control for Flexible Spacecraft

  Journal of Dynamic Systems Measurement and Control · 2026-05-07

  preprintOpen access

  Abstract Spacecraft are vital to space exploration and are often equipped with lightweight, flexible appendages to meet strict weight constraints. These appendages pose significant challenges for modeling and control due to their inherent nonlinearity. Data-driven control methods have gained traction to address such challenges. This paper introduces, to the best of the authors? knowledge, the first application of the data-enabled predictive control (DeePC) framework to boundary control for flexible spacecraft. Leveraging the fundamental lemma, DeePC constructs a non-parametric model by utilizing recorded past trajectories, eliminating the need for explicit model development. The developed method also incorporates dimension reduction techniques to enhance computational efficiency. Through comprehensive numerical simulations, this study compares the proposed method with Lyapunov-based control, demonstrating superior performance and offering a thorough evaluation of data-driven control for flexible spacecraft.

  PublisherDOI

• Learning solution operator of dynamical systems with diffusion maps kernel ridge regression

  ArXiv.org · 2025-12-19

  articleOpen access

  In this work, we propose a simple kernel ridge regression (KRR) framework with a dynamic-aware validation strategy for long-term prediction of complex dynamical systems. By employing a data-driven kernel derived from diffusion maps, the proposed Diffusion Maps Kernel Ridge Regression (DM-KRR) method implicitly adapts to the intrinsic geometry of the system's invariant set, without requiring explicit manifold reconstruction or attractor modeling, procedures that often limit predictive performance. Across a broad range of systems, including smooth manifolds, chaotic attractors, and high-dimensional spatiotemporal flows, DM-KRR consistently outperforms state-of-the-art random feature, neural-network and operator-learning methods in both accuracy and data efficiency. These findings underscore that long-term predictive skill depends not only on model expressiveness, but critically on respecting the geometric constraints encoded in the data through dynamically consistent model selection. Together, simplicity, geometry awareness, and strong empirical performance point to a promising path for reliable and efficient learning of complex dynamical systems.

  PublisherOA PDF

• Multi-objective Optimization of Rotorcraft Blade Structure with Multi-disciplinary Constraints

  2025-05-20

  articleSenior author

  We extend the previously developed integrated VABS (iVABS) framework for rotor blade structural optimization with an enhanced cross-section template for practical manufacture considerations; these include the introduction of curved spar corners, a continuous wrap-around skin, trailing-edge tabs and a conformal non-structural mass. The added fidelity is exercised on a UH-60A-based outer mold line through three multi-objective optimization case studies, including a case where the cross-sections are optimized independent of each other, and two cases where all the cross-sections are optimized simultaneously with manufacture considerations. It was found that the latter cases produce straight spars that are relatively more practical to manufacture when compared to the first case, while achieving significant reduction of up to 80% in the mismatch of stiffness values, inertia properties, and shear center locations, when compared to the prior work. A subsequent sensitivity analysis of the Pareto set isolates five critical combinations of design variables, out of the original nearly 100 variables, such as root/mid-span skin-ply count and spar-web placement, that can be adjusted to refine the Pareto solution. The study demonstrates that manufacturability-aware parameterization and data-driven variable reduction can deliver practical composite-blade designs within a scalable optimization loop.

  PublisherDOI

• Reduced-Order Modeling of Turbulent Flows for High-Speed Aerothermoelastic Analysis

  AIAA Journal · 2025-12-09 · 1 citations

  article

  This work presents a reduced-order model that efficiently generates unsteady aerothermal loads due to turbulent boundary layers for high-speed aerothermoelastic analysis. This prediction capability enables high-fidelity aerothermoelastic and structural fatigue analyses over long durations with tractable computational costs. Unsteady pressures over deformed structures are modeled by decomposing a turbulent boundary layer into temporal and spatial components. An unsteady pressure history over an undeformed structure is first reconstructed by superimposing spectral proper orthogonal decomposition modes and frequencies. Subsequently, the reconstructed pressures receive spatial corrections to model regions of flow compression and expansion. Heat fluxes over deformed structures are generated using a superposition/interpolation method that applies spatial corrections to the heat fluxes of an undeformed structure. An assumption required to construct the reduced-order model is verified: spatial pressure fluctuations and heat fluxes are a linear function of the modal coordinates of structural deformations in the linear regime. The reduced-order model produces accurate spectral contents of pressure fluctuations, which is critical for accurate structural excitation prediction; however, the model slightly overpredicts the fluctuation magnitudes. Likewise, the model produces accurate steady heat flux loads. The results enable high-fidelity unsteady aerothermoelastic simulations at a computational cost reduction of five orders of magnitude.

  PublisherDOI

• Global Description of Flutter Dynamics via Koopman Theory

  2025-01-03 · 1 citations

  articleSenior author

  We introduce a data-driven method for flutter analysis and prediction based on Koopman theory. The Koopman formalism enables the representation of nonlinear dynamics in a higher-dimensional linear space through the lifting of coordinates. The resulting linear model is valid over a broad region, and in some cases, globally, within the state space, offering a powerful tool for extending classical linearized stability analysis to a global stability assessment for flutter. In this paper, we present a extended bilinear model parameterized by flutter parameter to capture nonlinear behavior of flutter dynamics. We then establish a rigorous connection between the eigenvalues and eigenvectors of the extended bilinear model and those of the nonlinear flutter dynamics, addressing both fixed-point (equilibrium) and limit-cycle (flutter) cases. Finally, the proposed methods are applied to a 2D academic example and a more realistic panel flutter problem, highlighting how pre-flutter data can be utilized to characterize the flutter mechanism and predict the flutter boundary in a model-free, data-driven manner.

  PublisherDOI

• Low-Order $\mathcal{H}_2 / \mathcal{H}_\infty$ Controller Design for Aeroelastic Vibration Suppression

  ArXiv.org · 2025-12-11

  preprintOpen access

  This paper presents an $\mathcal{H}_2 / \mathcal{H}_\infty$ minimization-based output-feedback controller for active aeroelastic vibration suppression in a cantilevered beam. First, a nonlinear structural model incorporating moderate deflection and aerodynamic loading is derived and discretized using the finite element method (FEM). Then, a low-order linear model is identified from random gaussian input response data from the FEM model to synthesize an output-feedback controller using the $\mathcal{H}_2 / \mathcal{H}_\infty$ framework. A frequency-weighted dynamic filter is introduced to emphasize disturbance frequencies of interest, enabling the controller to target dominant vibration modes. Simulation results demonstrate the effectiveness of the proposed technique for vibration suppression and study its robustness to system parameter variations, including actuator placement.

  PublisherOA PDFDOI

• Aeroservoelastic Modeling for Trajectory Optimization of Morphing Aircrafts

  2025-05-05 · 1 citations

  articleSenior author

  Abstract Morphing aerial vehicles exhibit enhanced maneuverability when compared to their fixed configuration counterparts; this improves their mission performance, e.g., in obstacle avoidance. However, the trade-off in enhanced performance is the complexity of the controller design due to the increased degrees of freedom. This paper presents a mid-fidelity aeroservoelastic model for a morphing wing aircraft together with a trajectory optimization framework. As a benchmark problem, the flexible morphing wing aircraft is compared against the flexible non-morphing aircraft for a pure pull-up scenario, considering aeroservoelastic effects across three cases: non-morphing, uniform morphing, and independent morphing. The results show that symmetric morphing inputs improve terminal height in a pure pull-up maneuver by 9.9% over a non-morphing configuration, demonstrating significant benefits in maneuverability; meanwhile allowing independent morphing moments provides no additional performance gain. A gradient-free optimal trajectory planner based on Model Predictive Path Integral (MPPI) was also employed to show that the MPPI algorithm is a promising solution for trajectory optimization in an aeroservoelastic setting. The results validate the framework’s effectiveness for trajectory optimization with aeroservoelastic dynamics. Future work will focus on incorporating more accurate morphing control inputs and control costs into the optimization process and scaling up the framework for higher-dimensional control inputs in more complex flight missions.

  PublisherDOI

• Learning coarse-grained dynamics on graph

  Physica D Nonlinear Phenomena · 2025-06-18 · 2 citations

  articleCorresponding
  PublisherDOI

• Reduced-Order Modeling of Resolved Turbulent Flows for High-Speed Fluid-Structure Interaction Analysis

  2025-01-03

  article

  This study presents a reduced-order model method capable of generating resolved turbulent boundary layer pressure fluctuation loads at a reduced computational cost for aeroelastic analysis. The model generates unsteady pressure loads over a deformed panel by decomposing turbulent boundary layer flow into temporal and spatial components. First, an unsteady pressure fluctuation history over a flat plate is reproduced by superimposing spectral proper orthogonal decomposition modes and frequencies, which are orthogonal in space and time. Subsequently, spatial corrections to the pressure fluctuation magnitudes of the reconstructed flow are implemented to account for regions of flow compression and expansion over the deformed panel. Two assumptions required to construct the reduced-order model are verified: (1) one-way coupling exists between turbulence and structural responses, in that the impact of the former on the latter is negligible, and (2) spatial turbulent pressure variation is a linear function of modal coordinates of structural deformation. The reduced-order model produces accurate spectral properties of pressure fluctuations, which is critical for accurate prediction of the excitation of structural modes; however, the model overpredicts the magnitude. The results enable high-fidelity unsteady aeroelastic simulations at a reduced computational cost.

  PublisherDOI

Frequent coauthors

• Matteo Filippi

  Polytechnic University of Turin

  49 shared

• Cal Poly Pomona

  Pennsylvania State University

  49 shared

• Zahra Sotoudeh

  Oklahoma State University Oklahoma City

  49 shared

• Weihua Su

  49 shared

• Jinwei Shen

  49 shared

• Marco Petrolo

  Polytechnic University of Turin

  49 shared

• Yi Wang

  SAIC Motor (China)

  49 shared

• Wei Zhao

  Shanghai Jiao Tong University

  49 shared

Labs

• APUS LabPI

  Multi-Disciplinary Analysis, Optimization, and Control

Education

• Doctor of Philosophy, Department of Aerospace Engineering

  University of Michigan

  2019