space
Apr 11, 2025
Di Wu: Quantifying Debris and Advancing AI to Secure Earth’s Orbital Commons
An exploration of how artificial intelligence and economic modeling are being used to quantify, predict, and manage space debris, highlighting the integration of large language models and real-time simulation tools for smarter orbital decision-making.

Dr. Di Wu
As the volume of satellites orbiting Earth continues to surge, understanding and managing the resulting congestion has never been more critical. From quantifying the economic impact of fragmentation events to harnessing artificial intelligence for on-orbit decision making, Dr. Di Wu’s work sits at the forefront of space sustainability.
EcoAero had the opportunity to speak with Dr. Di Wu, Assistant Professor at Embry-Riddle Aeronautical University and former MIT postdoctoral researcher, about his pioneering research in space situational awareness, debris quantification, and the integration of large language models into astrodynamics.
Dr. Wu began by reflecting on his PhD work at UCSD, where he first tackled space debris characterization using machine learning techniques. During his MIT postdoc, he focused on putting numbers to the problem: developing metrics to quantify the economic value of preventing fragmentation events, and assessing how long it would take for debris clouds to render orbital regions unusable.
Building on this quantitative foundation, Dr. Wu and collaborators introduced the first ever astrodynamics problem benchmark for evaluating large language models.
“So one aspect of our innovation was actually kind of like directly qualifying the final results. So new map, we have both numerical results and analytical results.”
Beyond simple yes-no answers, this benchmark provides both analytical solutions and numerical accuracy metrics—complete with error buffers (e.g., a 10% boundary for 99% of problems) that can tighten as models improve, offering actionable guidance for future space engineering tasks.
To capture the influence of human activity on orbital debris evolution, Dr. Wu devised the LRCI (Learned Reality Check Index). By comparing long-term debris simulations from the MIT ARC Lab’s MOCAT toolbox against historical launch and fragmentation data dating back to the 1960s, his team quantified how events like the 2020 launch slowdown during the pandemic temporarily altered the debris environment.
With mega-constellations now launching hundreds of satellites each month, Dr. Wu highlights that industry is already optimizing mission designs for sustainability. Yet new services—such as on-orbit refueling or inspection—bring both opportunities and risks.
“Some of the future challenge I see is actually on this, like new technology emerging and new service emerging, and how do we incorporate those things into the broad picture about space sustainability? … What is going to be the quantified benefit versus the quantified risks?”
Looking ahead, Dr. Wu sees “efficiency” as the key to sustainable space operations. Rather than preserving vast empty corridors, the goal should be to maximize the use of existing orbital “real estate” while ensuring safe, smooth traffic flow—principles that will extend from Earth-orbit into cislunar space.
Real-time debris detection and avoidance will depend on integrating advanced sensors—both ground-based and onboard satellites—with powerful AI systems. As Dr. Wu explains, large language models could translate vast streams of sensor data into actionable collision-prediction insights, while simulation platforms like MOCAT can vet and validate optimal traffic-management policies before they’re implemented in orbit.
Although his research has been rooted in academia, Dr. Wu stresses that commercialization is vital for progress. His XD Lab at Embry-Riddle is actively collaborating with startups to embed intelligent debris-mitigation capabilities directly onboard satellites, bridging the gap between theory and operational systems.
Dr. Wu’s work illustrates the power of combining rigorous quantification, cutting-edge AI benchmarks, and industry partnerships to tackle one of the most pressing challenges in aerospace today. His lab remains open to visitors and student researchers—an invitation for the next generation to join the mission of preserving our shared orbital commons.
At EcoAero, we’re committed to spotlighting visionary research like Dr. Wu’s that marries rigorous quantification with AI-driven solutions to safeguard our shared orbital commons. We will continue to amplify these breakthroughs, foster collaboration between academia and industry, and equip the next generation of space stewards to ensure a sustainable, efficient future in Earth-orbit and beyond.
More in
space

space
Feb 9, 2024
Obruta Space: How AI-Powered Pilots Are Enabling the Orbital Economy
Insights into how autonomous technology and artificial intelligence are transforming orbital infrastructure through cost-effective spacecraft refueling, maintenance, and debris removal systems.

space
Apr 17, 2025
Nilton Renno: Harnessing Hypervelocity Signals to Detect the Invisible Threat of Space Debris
A detailed look at innovative detection methods for small-scale space debris using electromagnetic signals from hypervelocity collisions, enhancing orbital safety through improved situational awareness.