AE Seminar

"Accelerating Aerospace Engineering Design with Large Language Model Driven Exploration and Reasoning" 

featuring

Ping He

Assistant Professor of Aerospace Engineering | Iowa State University

Thursday, April 9

11:00 a.m. - 12:00 p.m.

Guggenheim 442

About the Seminar:

In this talk, I will present a new paradigm for large language model (LLM)-driven multidisciplinary design optimization (MDO). The framework builds AI agents that combine deterministic tool execution with heuristic design reasoning, enabling fully conversational, end-to-end MDO workflows. The first part of the talk introduces our scalable MDO framework, which integrates high-fidelity solvers for aerodynamics, structures, and heat transfer. We develop efficient discrete adjoint algorithms to enable gradient-based optimization with hundreds of design variables and constraints, making large-scale MDO practical. A range of MDO applications, including aircraft, propellers, and thermal systems, will be presented to demonstrate the framework’s versatility and robustness. The second part focuses on integrating this MDO framework with LLM-driven AI agents to enable a conversational design workflow. These agents orchestrate geometry manipulation, mesh generation, simulation, optimization, and post-processing in a deterministic and verifiable manner, while simultaneously performing high-level design reasoning. We will demonstrate how AI agents can iteratively adjust MDO formulations, such as changing design variables, objectives, and constraints, to better explore design spaces and tradeoffs, achieve more practical designs, and facilitate decision making. Examples include wing aerostructural optimization, propeller aeroacoustic optimization, and UAV trajectory optimization. This paradigm has the potential to broaden access to advanced MDO capabilities and significantly accelerate aerospace engineering design. 

About the Speaker:

Dr. Ping He is an Assistant Professor of Aerospace Engineering at Iowa State University. Prior to joining ISU, he conducted postdoctoral research at the University of Michigan and at North Carolina State University. He earned his Ph.D. from the Chinese Academy of Sciences. His research focuses on integrating emerging agentic AI with MDO to accelerate the design of high-performance aerospace systems, including aircraft, turbomachinery, and spacecraft.