About the Challenge
The ADL Design Challenge is an annual, intensive engineering program hosted by Stanford’s Aerospace Design Laboratory in San Francisco. Individual participants - engineers, researchers, and graduate students from around the world - come together on site, form teams, and work side-by-side on real fixed-wing UAV design problems.
What We Do
The challenge emphasizes the process of design: from mission definition through aerodynamic parametrization, AI-driven optimization, rapid prototyping, and finally flight validation. Participants leave with hands-on experience in the same computational tools and methods used in the aerospace industry.
How Teams Work
Teams are formed on arrival - mixing backgrounds, institutions, and skill sets. You might work alongside a propulsion specialist from Munich, a controls engineer from Tokyo, or a CFD researcher from Sao Paulo. That cross-pollination of perspectives is the point.
Every edition takes place in San Francisco, hosted on site at Stanford. Past cohorts have tackled challenges ranging from high-endurance surveillance platforms to agile mapping UAVs - always with a focus on doing the engineering right, not just getting to an answer fast.
The Design Process
1. Define & Parametrize
Establish mission requirements, define the flight envelope, and build a parametric model linking geometry to performance. Every design variable is quantified and linked to the aircraft’s behavior.
2. Optimize
Train neural network surrogate models on simulation data and run optimization algorithms to explore the design space. Replace expensive CFD runs with fast predictions, then converge on the best configuration.
3. Build & Validate
Rapid-prototype the design using 3D printing, laser cutting, and foam construction. Integrate propulsion, electronics, and control systems. Fly test missions and compare real data back to predictions.
Who Should Apply
The challenge is open to graduate students, early-career researchers, and engineers with a background in aerospace, mechanical engineering, computer science, or related fields. You don’t need UAV experience - we teach the tools and methods. What matters is curiosity, technical rigor, and the willingness to learn by doing.
Tools & Methods
- XFLR5 and OpenVSP for aerodynamic analysis
- Python-based neural network surrogate modeling
- Genetic algorithms and gradient-based optimization
- CAD parametrization and rapid prototyping
- Flight data logging and post-flight analysis
- SU2 for CFD and shape optimization
- SUAVE for conceptual aircraft design