About

James Baeder is the Igor Sikorsky Distinguished Professor in Rotorcraft at the Department of Aerospace Engineering at the University of Maryland. He holds a B.S. in Mechanical Engineering from Rice University, and both an M.S. and Ph.D. in Aeronautics and Astronautics from Stanford University. His research interests focus on computational aerodynamics and aeroacoustics algorithms, with specific attention to coupling computational fluid dynamics (CFD) to comprehensive rotor analysis. He is also interested in using CFD to generate training data for machine-learned surrogate models and employing such models for prediction and inverse design. Professor Baeder has made significant contributions to the field through his work on high-resolution wake capturing methodologies for hovering rotors, rotor aerodynamics, and the analysis of vortex-dominated flows. He has been actively involved in the aerodynamics and propulsion community, serving on various technical committees and as a reviewer for the American Helicopter Society and the AIAA. His scholarly work includes numerous journal articles and conference presentations that advance the understanding of rotorcraft aerodynamics, wind tunnel wall interference effects, and the aerodynamics of micro hover rotors. Baeder's leadership in rotorcraft research is recognized through his promotion to Professor and his role as the Alfred Gessow Rotorcraft Center's faculty member, supporting research in rotorcraft technology and innovation.

Research topics

• Mechanics
• Physics
• Optics
• Geometry
• Mathematics
• Engineering
• Structural engineering
• Classical mechanics

Selected publications

• Generalizable Deep Learning Module for Rotorcraft Inverse Design Applications

  Journal of Aircraft · 2026-03-09

  articleSenior author

  Airfoil design optimization involves iterative processes, often using less accurate low-fidelity tools due to limitations on computational costs. Surrogate modeling bridges the gap between low computational costs and high-accuracy analysis tools (such as computational fluid dynamics). This study explores the benefits of a vectorized implementation of artificial neural networks (ANNs) as a surrogate modeling tool to predict the performance polar (profiles of lift, drag, and pitching moment coefficients) of airfoils at various angles of attack with high accuracy and a more uniform error distribution. The ANNs are additionally trained on the boundary-layer thickness at the airfoil’s trailing edge. For inverse design challenges, we introduce the tandem neural network (T-NN) architecture, a novel approach utilizing the entire performance polar instead of a single point. Comprising two ANNs working in series, T-NNs feature a modified and flexible loss function, thereby enhancing accuracy in inverse-design problems. Practical design constraints are seamlessly integrated into the design loop, making this architecture a valuable tool for designers. To illustrate real-world applicability, we include the implementation of this deep learning framework for rotor blade design, showcasing the versatility and effectiveness of the proposed methodology.

  PublisherDOI

• Functional Correction to Spalart–Allmaras Model for Adverse Pressure Gradient and Low-Reynolds Number

  AIAA Journal · 2026-03-18

  articleSenior author

  Turbulence modeling remains a critical challenge in computational fluid dynamics, particularly when simulating complex flow regimes. While the Spalart–Allmaras (S-A) turbulence model is widely regarded for its computational efficiency and robustness in attached flows with favorable pressure gradients, its predictive accuracy deteriorates under adverse pressure gradient (APG) conditions—such as flow separation and stall—and at low Reynolds numbers (Re), where skin friction coefficient predictions exhibit significant discrepancies in regions of low turbulent Reynolds number, [Formula: see text], within the developing boundary layer. In prior work, we introduced an adverse pressure gradient correction (APGC) that substantially improved lift coefficient predictions in APG-dominated flows without compromising drag coefficient accuracy. Concurrently, an existing low-Reynolds-number correction (LRe) demonstrated enhanced fidelity in skin friction predictions at low Re. Building upon these findings, the present study proposes a unified framework integrating both APGC and LRe corrections into the S-A model. This synergistic modification aims to extend the model’s applicability, improving predictions of lift, drag, and incipient angle of attack across a broader range of flow conditions with negligible computational overhead. The outcome is a refined S-A model capable of addressing key limitations in APG and low-Re regimes while nearly maintaining the original formulation’s efficiency.

  PublisherDOI

• Machine Learning Framework for Predicting Shock Locations on Rotorcraft Airfoils

  2025-07-16 · 1 citations

  articleSenior author

  The rotor blade tip can operate at the transonic regime and location of transonic shock is important. Most mid/lower fidelity comprehensive codes for rotor blade analysis uses airfoil tables. The airfoil tables are usually generated using Computational Fluid Dynamics (CFD) codes, and the it's crucial to capture the tansonic shock to get accurate integrated quantities. This study focuses on predicting the location of transonic shocks on rotorcraft airfoils using a machine-learning-based surrogate model. A comprehensive methodology is developed to parameterize airfoil geometries, generate a design of experiments, and train neural network models for predicting shock locations. CFD simulations are performed on 560 unique airfoils using an in-house solver (HAM2D) over a range of angles of attack at Mach 0.7 to obtain a large sample of shock data. A combination of classification and regression models are employed to detect the presence of shocks and predict the location of shocks. The models demonstrate high accuracy, effectively predicting shock occurrence and locations, with regression errors well within acceptable tolerances for localized mesh refinement. The proposed framework provides a robust and efficient approach for predicting shock locations, which is essential for optimizing rotorcraft performance under transonic conditions.

  PublisherDOI

• A Comprehensive Review of Neural Network Training Approaches for Airfoil Design and Optimization

  2025-01-03 · 7 citations

  reviewSenior author

  Airfoiloptimizationisthepreliminarystepinanyaerodynamicdesignoptimizationproblem. The use of computational fluid dynamics (CFD) codes for aerodynamic simulations and optimization is computationally expensive. There have been recent advances in surrogate modelingtechniquesforthepredictionofaerodynamiccoefficientsofinterest(liftcoefficient (����), drag coefficient (����) and pitching moment coefficient (����)) for a specified airfoil geometry. These surrogate models are trained not only to predict the performance polar (����,����, and����) but also for inverse design aimed at achieving a specific target performance and for facilitating designoptimization. Thispaperprovidesadetailedstudyofneuralnetworkframeworksfor airfoilpolarpredictionandinversedesign. Theaspectsofgeometryparameterization,neural network training, inverse design architecture and data reduction strategy are also presented in this paper

  PublisherDOI

• Eigenvalue Sensitivity Computations for Linear Stability Theory

  Journal of Aircraft · 2025-02-10

  articleSenior author

  To realize the drag reduction benefit of boundary-layer transition control strategies, it is crucial to integrate transition prediction into the vehicle design through an optimization process. The integration of transition prediction based on linear stability analysis into adjoint-based design optimization requires coupling an adjoint-enabled computational fluid dynamics (CFD) solver with an adjoint-enabled linear stability code. In particular, the boundary-layer transition location is often predicted using the [Formula: see text]-factor method based on linear stability theory (LST). Thus, the sensitivity of the linear-stability eigenvalues constitutes an essential building block for optimizing the laminar flow performance. The present paper describes an implementation of LST eigenvalue sensitivity analysis that can be easily coupled with a CFD solver. Specifically, we describe a discrete adjoint formulation for the transition location prediction based on the [Formula: see text]-factor method. The verification of this formulation is carried out by comparing the adjoint-based sensitivity of the local growth rate of a given instability mode with respect to the disturbance frequency and the adjoint-based sensitivity of the transition location with respect to the spanwise wavenumber with those sensitivities computed using a finite-difference approximation. Finally, the adjoint LST formulation is applied to flat-plate boundary-layer flows at transonic, supersonic, and hypersonic conditions to determine the behavior and sensitivities of the transition location with respect to a range of disturbance spanwise wavenumbers.

  PublisherDOI

• Computational Analyses of Coaxial Rotor Hub Flows and Correlation with Experiment

  2025-05-20

  article

  This study presents computational analyses of coaxial rotor hub flows and validation against experimental data obtained from the fifth Rotor Hub Flow Prediction Workshop. Experiments were conducted in a 12-inch diameter water tunnel at Pennsylvania State Applied Research Laboratory, employing tomographic particle-image velocimetry (Tomo-PIV) and precise hub drag measurements. Three CFD codes (UMD Mercury, CREATETM-AV Helios, and OVERFLOW) utilizing hybrid Reynolds-Averaged Navier-Stokes (RANS) / Large Eddy Simulation (LES) modeling based on Spalart–Allmaras turbulence model, were applied to replicate and analyze hub flows. Counter-rotating coaxial rotor hubs under free-air condition was simulated as the simplest case and the hub drags are compared between the three CFD codes. The full water tunnel configuration, consisting of two hubs, a fairing, and shafts, was also simulated and compared to experimental results, with a focus on hub drag, wake velocity fields, and turbulence quantities. Results demonstrated that the computational frameworks effectively captured key flow physics, although some discrepancies in drag harmonics, wake velocity and turbulence intensity magnitudes were observed. Additionally, the study highlighted the impact of rotor hub geometry and installation of sail-fairing on drag and wake structures. These findings contribute to improve computational predictions, essential for designing high-speed rotor hub configurations.

  PublisherDOI

• Multi-fidelity Analysis of Quadrotor Biplane Tailsitter Hover-to-Cruise Transition

  2025-01-03

  articleSenior author

  The versatility of unmanned aerial vehicles (UAVs) has resulted in a large recent demand for their development, these aircraft can be useful in a variety of fields such as package delivery, search and rescue, agriculture, recreation and others. In particular, quadrotor biplane tailsitter configurations have the potential to become increasingly popular due to their efficiency in both hover and forward flight, one of these aircraft is the CRC-10 developed at the Army Research Lab (ARL). This study focuses on investigating the best approach to performing accurate hover-to-cruise transition flight simulations of a scaled up, 80 lbs, pitch controlled CRC-10. A flight dynamics and controls code is coupled to both low and mid-fidelity aerodynamics to generate a precise, force balanced transition flight simulation.

  PublisherDOI

• Aerodynamic Analysis of Stopped and Stopping Rotors in Lift+Cruise eVTOL Configurations

  2025-05-20

  article

  This study investigates the aerodynamic behavior of lift rotors in a representative lift+cruise electric vertical takeoff and landing (eVTOL) configuration using high-fidelity Computational Fluid Dynamics (CFD) simulations. As lift+cruise concepts gain prominence for Urban Air Mobility (UAM) applications due to their operational simplicity, flight performance, and reduced cruise noise, a detailed understanding of rotor aerodynamics during transition and cruise is critical. CFD analysis was conducted for both slowed rotors at high advance ratios and fully stopped rotors, where traditional predictive tools become inaccurate. Results show that lift rotors operating at advance ratios approaching three exhibit quasi-steady behavior similar to stopped rotors. The influence of rotor lock orientation on aerodynamic loads was characterized, with a freestream-aligned lock angle minimizing drag and asymmetry. A rotor hub fairing was found to reduce blade root separation and drag, though at the cost of slightly increased hub moments. The sensitivity of axial loads to vehicle pitch angle was quantified, highlighting the need to account for effective angle of attack in cruise load predictions. These findings inform future modeling strategies, control system design, and airframe integration for advanced eVTOL vehicles concepts.

  PublisherDOI

• Prediction of Hover Validation and Acoustic Baseline Hovering Rotor Performance and Ground Effect Using Mercury Framework

  Journal of Aircraft · 2025-02-10 · 2 citations

  articleSenior author

  Simulations of the Hover Validation and Acoustic Baseline rotor in the hovering state were conducted using the Mercury framework, a multimesh, heterogeneous CPU-GPU overset computational fluid dynamics framework. Isolated hovering rotor simulations were conducted, and efficiency parameters were compared with the available reference data for various collective pitch angles. For installed rotor simulations, the NASA ROBIN-Mod7 fuselage was used. The sectional thrust of the blade increased when passing over the nose and tail boom of the fuselage owing to the ground effect, whereas the fuselage download decreased the total lift. The hover-in-ground effect was investigated for the rotor–fuselage configuration at various rotor heights from the ground. The ground effect increased as it approached the ground, thereby increasing the rotor figure of merit and decreasing fuselage download. The sectional thrust of the blade increased under the ground effect, except near the tip. Otherwise, the sectional torque was not significantly affected. Finally, the groundwash velocity was computed for an installed rotor at a collective of [Formula: see text]. Owing to the highly unsteady flow features, the results were phase- or time-averaged. The ground velocity profiles at various azimuthal and radial locations were compared to study the effects of rotor height, fuselage existence, and ground-wall boundary conditions.

  PublisherDOI

• INTEGRATE – Inverse Network Transformations for Efficient Generation of Robust Airfoil and Turbine Enhancements

  2025-06-05

  report1st authorCorresponding
  PublisherDOI

Frequent coauthors

• Yong Su Jung

  Pusan National University

  34 shared

• W. J. Mccroskey

  28 shared

• Bumseok Lee

  University of Maryland, Baltimore

  19 shared

• Karthikeyan Duraisamy

  University of Michigan–Ann Arbor

  17 shared

• Shivaji Medida

  Altair Engineering (United States)

  16 shared

• Vinod K. Lakshminarayan

  16 shared

• Camli Badrya

  University of California, Davis

  13 shared

• Jayanarayanan Sitaraman

  United States Army Combat Capabilities Development Command

  12 shared

Awards & honors

• Igor Sikorsky Distinguished Professor in Rotorcraft
• Fellow of the American Helicopter Society
• Technical Fellow of the American Helicopter Society
• Zonta International Amelia Earhart Fellowships
• Dean's Doctoral Research Awards