Workshop on AI in Aeronautical Engineering: Intelligent Methods for Aircraft Design, Simulation and Optimization

Scope and Objectives

This workshop deals with how modern AI methods, especially large language models (LLMs), physics-informed neural networks (PINNs), and other data-driven frameworks, can revolutionize the aircraft engineering and system design lifecycle. The broad and sudden availability of LLMs have left a gap between the capability of the AI-tools and their application in the engineering disciplines, which leaves a wide open field of opportunity for domain experts, for research and development into the application of these tools into engineering processes.

We welcome work on the overall implications of AI in complex systems and AI methods in engineering design work, both conceptual design and detailed engineering, including integrating AI into computational workflows (e.g., CFD, FEA), design-space exploration, and rapid prototyping. The aim is to highlight the characteristics of complex systems with integrated AI solutions, including tools and networks that embed physical laws and domain knowledge directly within learning architecture. In addition, the workshop will address aspects of how these developments interface with aeronautical engineering education and computational AI-infrastructure, such as computers for high-fidelity simulation and optimization. There will be ample possibilities for interaction, discussions and networking, e.g. panel discussion on relevant topics aimed at exchange of ideas.

The workshop is by invitation only. If there is an interest contact one of the organisers.

Topics of Interest

Potential topics include, but are not limited to:

• AI and the emergence of complex intelligent systems: Changing characteristics of complex systems and consequences for engineering design and management
• AI and engineering design: Overall implications of the integration of AI methods in engineering design of complex systems. Explainability and Auditability of AI methods in Engineering and Aeronautics industry.
• Physics-Informed & Hybrid Modeling: Embedding physical laws and domain knowledge within AI frameworks to enhance simulation accuracy and efficiency.
• Large Language Models as Design Assistants: AI-tools for automated engineering document generation (requirements,specs, reports), and for concept generation and modelling for system analysis and evaluation.
• Surrogate and Reduced-Order Modeling: Data‑driven metamodels built with modern deep‑learning architectures for design analytics
• Interactive & Generative Design: Deep generative models (VAEs, GANs, diffusion models) for e.g. novel geometries. NASA
• Optimization and control: How will AI tools be integrated in optimization and control of Aeronautical Engineering systems.
• Infrastructure: What are the implications of the modern large scale computation AI-infrastructure for Aeronautical Engineering.
• Education: How will the AI tools be integrated in the modern Aeronautical Engineering curriculum.

Organizing Committee

• Prof. Emilia Villani (ITA, Brazil)
• Prof. Dan Henningson (Royal Institute of Technology, Sweden)
• Prof. Petter Krus (Linköping University, Sweden)
• Prof. Fredrik Heintz (Linköping University, Sweden)
• Lt. Col. Guilherme Moreira (ITA, Brazil)
• Prof. Ana Carolina Lorena (ITA, Brazil)

Programme: