Cody Karcher "Using Optimization and Artificial Intelligence for Aircraft Design"

4:00 pm
Monday, January 22, 2024
190 Goodwin Hall
Faculty Host:  Dr. Rakesh Kapania

Abstract:  There is little doubt that computational tools have revolutionized aircraft design, and for the past 30 years a significant research effort has focused on how to most effectively utilize modern computational resources in the design process.  Researchers in the field of Multi-Disciplinary Analysis and Optimization 

(MDAO) have proposed many strategies for connecting existing analysis tools (CFD, FEA, etc.) together with an optimization algorithm to improve the performance of designs.  But recent literature has suggested that prioritizing optimization at the expense of analysis fidelity can result in a more effective utilization of computational resources.  At the same time, the recent and rapid expansion of interest in Generative Artificial Intelligence, specifically Large Language Models (LLMs) like ChatGPT, has opened up a new way of interacting with computational tools that could revolutionize the future of design.  In this talk, I will discuss lessons learned from MDAO and lay out a vision for how AI can be used to assist in the aircraft design process.

Bio:  Cody Karcher is the Computer Science Research Institute (CSRI) Postdoctoral Fellow at Sandia National Laboratories.  He completed his PhD at MIT (2022), which focused on developing optimization algorithms tailored to aircraft design applications, and did his undergraduate degree at the University of Maryland, College Park (2014).  Cody's research focuses on improving the aircraft design process through the use of computational engineering, applied math, and artificial intelligence.  Prior to his current position at Sandia he was a postdoctoral fellow at the University of Michigan, and has spent time at MIT Lincoln Laboratory, Boeing, and NASA.  He enjoys sailing, curling, and cycling in his free time.