About

Juan Alonso is the Vance D. and Arlene C. Coffman Professor and the James and Anna Marie Spilker Chair of the Department of Aeronautics and Astronautics at Stanford University. His research focuses on autonomous systems, controls, and aerospace design, contributing to advancements in these fields through his leadership and scholarly work. As a faculty member, he is involved in teaching and mentoring students, and his expertise supports the department's mission to innovate in aeronautics and astronautics.

Research topics

• Computer Science
• Engineering
• Aerospace engineering
• Physics
• Automotive engineering
• Simulation
• Mathematics
• Mechanics
• Reliability engineering
• Marine engineering
• Aeronautics
• Computational science
• Algorithm
• Programming language
• Mechanical engineering
• Thermodynamics
• Mathematical optimization

Selected publications

• A Comparison of Methodologies for Generating Design-Requirement-Driven Conceptual Aircraft Designs

  2026-01-08

  article

  Traditional aircraft design techniques leveraging high-fidelity analyses are often limited by the prohibitive cost of design space exploration. Therefore, in the context of large design spaces and unconventional aircraft concepts, the use of machine learning to propose suitable designs represents an attractive solution for conceptual-stage exploration, a process termed generative design. In this paper, we construct a large dataset of narrow-body commercial airliner configurations, which we use to compare several generative machine learning techniques - including variational autoencoders (VAEs), normalizing flows, and diffusion models - to assess their ability to generate diverse, feasible, and realistic conceptual aircraft designs in a fraction of the time required to perform a full conceptual design. Preliminary results indicate that normalizing flows and diffusion models generate the more representative and diverse designs than VAEs. This study highlights how machine learning can support conceptual aircraft design, with relevance to unconventional configurations and sustainable aviation.

  PublisherDOI

• Correction: A Comparison of Methodologies for Generating Design-Requirement-Driven Conceptual Aircraft Designs

  2026-01-12

  article
  PublisherDOI

• Uncertainty Quantification of Machine Learning Models With Adaptive Sampling Applications

  2026-01-08

  articleSenior author

  Surrogate models of physical processes enable fast predictions in engineering contexts, reducing the time and resources spent during the online phase of many-query applications, including uncertainty quantification and optimization, among others. However, their effective use in design, design under uncertainty, or even as part of certification-by-analysis processes, requires that their predictions include some indication of how much they can be trusted. This requirement has led to substantial advances in the field of uncertainty-aware surrogate models, which return estimates of their own predictive uncertainty along with their predicted values. We compare seven different uncertainty-aware surrogate models, including Gaussian processes, deep learning methods, and various types of conformal prediction, on the basis of their ability to generate accurate and informative estimates of their own uncertainty. Our results show that deep learning based methods significantly outperform Gaussian processes even in the low-data regime, and that conformal prediction can effectively correct over-confidence in deep-learning-based surrogates. In the aerospace industry, collecting data on which to train deep learning surrogate models can be expensive, so we additionally compare three strategies for constructing training datasets. We demonstrate that an adaptive sampling strategy, which selects training points using a novel gradient-and-uncertainty-based acquisition function, can enable aerospace surrogate models to achieve high accuracy with significantly less training data than random sampling. Overall, our results highlight that uncertainty-aware deep learning models have large practical potential to streamline aerospace workflows. To aid further research, documented implementations of all considered models are available at the UQRegressors Github Repository.

  PublisherDOI

• Multifidelity Rotor Wake Modeling with Uncertainty Quantification of Transitioning Tilt-Propeller Aircraft Performance

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

  article

  A framework for efficient multifidelity modeling of tilt-propeller aircraft performance is developed, with an emphasis on accurate modeling of transition maneuvers between vertical and forward flight. Low- and midfidelity vortex models are used for the propeller aerodynamic data sources and lifting surface analyses. Uncertainty quantification of propeller model parameters is conducted and used as the input uncertainty of the source data over the operational domain. A multifidelity approach is proposed, using active learning of Gaussian processes (GPs) that are sequentially and adaptively enhanced with additional higher-fidelity data queried at strategic points selected from an acquisition function. This process is applied to the buildup of single- and multifidelity GPs, forming surrogate models of propeller performance over the operational domain. Each surrogate model is used in the full aircraft mission analysis of a tilt-propeller aircraft and provides quantified uncertainty in aircraft performance over the full flight envelope. Optimal maneuvers are also explored, balancing the tradeoff between maximum power, tilt rates, and time efficiency of conversion. The developed methods are applicable to many distributed electric propulsion aircraft configurations and are applied in this paper to a pusher-configured tilt-propeller aircraft.

  PublisherDOI

• WITHDRAWAL: A Space-Time Collocation Method for Solving Burger's Equation via Convex Relaxation

  2025-07-29

  articleSenior author
  PublisherDOI

• Performance Evaluation of a Graph Neural Network-Augmented Multi-Fidelity Workflow for Predicting Aerodynamic Coefficients on Delta Wings at Low Speed

  2025-01-03 · 3 citations

  articleSenior author

  Conceptual design of a stable and efficient supersonic transport (SST) aircraft during takeoff and landing, where higher angle-of-attack (AOA) induces complex aerodynamic phenomena such as vortex lift and flow separation, can be challenging. Existing analysis methods face a fidelity-cost trade-off: high-fidelity (HF) methods, such as computational fluid dynamics (CFD), offer prediction accuracy but are computationally expensive for solver-in-the-loop analyses, while low-fidelity (LF) methods, such as vortex lattice method (VLM), lack the ability to capture nonlinear flow physics. To address this gap, this paper proposes a multi-fidelity conceptual design analysis workflow that integrates a graph neural network (GNN)-based surrogate model into VLM to augment the analysis fidelity of LF tools. The surrogate model learns the discrepancies between LF and HF pressure fields, enabling accurate and efficient aerodynamic analyses on arbitrary quantities of interest. When evaluated on a dataset with various Delta wing geometries, the proposed workflow achieves an approximately fivefold reduction in the normalized root mean square error (NRMSE) for the predicted lift, drag, and pitching moment coefficients compared to using VLM alone. The results also highlight the proposed workflow’s generalizability across new flow conditions and wing geometries, while identifying its limitations in prediction accuracy variance across the test dataset. Overall, the proposed workflow provides an efficient and effective framework for aerodynamic assessment in conceptual design with improved fidelity.

  PublisherDOI

• Large Eddy Simulation of the Early Jet-Vortex Interaction Phase of Contrails From Hydrogen Aircraft

  2025-07-16 · 1 citations

  article

  Contrails have recently gained widespread attention, as their estimated warming potential is in the same order as aviation’s CO2 and is largely more uncertain. Our research is motivated by concerns about future hydrogen-fueled aircraft, as they will emit considerably more water vapor, albeit no soot. In this study, we compare the ice crystal number and net radiative forcing of contrails forming behind kerosene- and hydrogen-fueled aircraft with large-eddy simulations (LES) of the jet-vortex interaction phases of contrails. We simulate the jet-vortex interaction using prescribed axial and vortical velocity fields and employ Lagrangian particle tracking and microphysical models for ice crystal formation and growth. We find that hydrogen contrails have a fraction to half the number of ice crystals of kerosene contrails, except when the engine operates in the soot-poor regime and flies at lower altitudes. Higher atmospheric aerosol concentrations may also increase the net radiative forcing of contrails from hydrogen-fueled aircraft with respect to an equivalent kerosene-fueled aircraft.

  PublisherDOI

• WITHDRAWN: A Space-Time Collocation Method for Solving Burger's Equation via Convex Relaxation

  2025-07-16

  articleSenior author

  The ability to solve partial differential equations accurately, robustly, and scalably is a core aspect of computational fluid dynamics and, by extension, computational aerospace shape optimization. In this paper, we investigate a novel approach to the solution of nonlinear PDEs in conservation form. In particular, we consider a spectral approach where the complete space--time solution is approximated as a polynomial-augmented network of radial basis functions. Our approach extends Kansa's method to nonlinear PDEs in conservation form by introducing a convex relaxation for the flux term. Although the approach is quite general, we demonstrate its efficacy in solving the one-dimensional viscous Burger's equation. In addition to being meshless, this approach allows future exploration of one-shot convex optimization approaches to aerodynamic shape optimization.

  PublisherDOI

• GPU-Based and Adaptive Solution Technology for the 5th AIAA High Lift Prediction Workshop

  2025-01-03 · 1 citations

  articleSenior author

  The Fifth AIAA Computational Fluid Dynamics (CFD) High Lift Prediction Workshop (HLPW-5) was organized to assess the numerical prediction and physical modeling capabilities of CFD technology for swept wings in landing and takeoff high-lift configurations. These calculations stress the geometry, meshing, modeling, solver, visualization, and post-processing aspects of CFD. Graphics Processing Units (GPUs) provide a step increase in computational performance but can also increase the complexity and time required to develop, deploy, and execute the CFD pipeline in a traditional on-prem scenario. Cloud computing provides scalable and elastic resources, which mitigate capacity and lead time constraints for the simulation engineer. By using modern software development and deployment processes to deliver GPU-based simulation technology via the cloud, CFD simulation as a service (SimaaS) is enabled, reducing the time from geometry to insight, a key element of the 2030 CFD Vision Study. In this paper, we provide a high-level description of the Luminary solver discretization and solution approach for compressible, turbulent flows, as well as the required mapping to modern GPU architectures. We also describe the Luminary geometry and meshing pipeline and our automatic mesh adaptation approach. Verification and Validation (V&V) results are then presented for \HLPW5 Cases 1 and 2. Solutions are computed on both committee-supplied meshes and on meshes automatically generated with Luminary technology. In addition, we exercise Luminary mesh adaptation (LMA) capabilities and demonstrate that solution-adaptive meshing greatly improves convergence to the correct solution with lower control volume count and in a fully automated way. Finally, performance and scalability benchmarks are presented on modern GPU architectures, and we estimate future performance metrics on next-generation GPUs.

  PublisherDOI

• Graph Neural Network-Guided Aerodynamic Shape Optimization for Conceptual Design of Supersonic Transport Wings

  2025-07-16

  articleSenior author

  Low-speed stability constraints have historically been difficult to incorporate during the conceptual design phase due to the limited accuracy of low-fidelity simulation methods. This paper presents a graph neural network (GNN)-guided multi-fidelity optimization framework for the aerodynamic shape design of supersonic transport wings at the conceptual level. A GNN-based field prediction surrogate model is integrated into the SUAVE aircraft design environment, enabling real-time, high-fidelity-informed aerodynamic shape optimization across a defined design space. Using a two-point, stability-constrained shape optimization problem, we demonstrate that the GNN-guided workflow produces a design that satisfies both stability and aerodynamic performance requirements, while the traditional vortex lattice method (VLM)- based workflow results in a configuration that violates the intended design objectives. The proposed approach improves the fidelity of aerodynamic analysis and enhances confidence in conceptual design, increasing the likelihood that the resulting configuration will perform well in later design phases.

  PublisherDOI

Recent grants

• CyberSEES: Type 1: Collaborative Research: Large-Scale, Integrated, and Robust Wind Farm Optimization Enabled by Coupled Analytic Gradients

  NSF · $200k · 2015–2017

Frequent coauthors

• Thomas D. Economon

  61 shared

• Francisco Palacios

  Universitat Politècnica de Catalunya

  45 shared

• Antony Jameson

  38 shared

• A. Jameson

  Texas A&M University

  24 shared

• Edwin van der Weide

  24 shared

• Brendan Tracey

  17 shared

• Andrés Santiago Padrón

  Stanford University

  16 shared

• Seongim Choi

  Gwangju Institute of Science and Technology

  16 shared

Education

• Ph.D., Aeronautics and Astronautics

  Stanford University

  1994

• M.S., Aeronautics and Astronautics

  Stanford University

  1989

• B.S., Aeronautical Engineering

  University of California, San Diego

  1986

Awards & honors

• AIAA: Excellence in Teaching Award
• AIAA: Outstanding Course Assistant
• William F. Ballhaus Prize
• Cannon Summer Fellowship
• Hoff Outstanding Master’s Student