New study quantifies uncertainty in space debris capture using tether-net systems
Noisy sensors could slash success rates of net-based debris capture in orbit.
As Low Earth Orbit grows crowded with debris, active tether-net systems with maneuverable units offer a promising removal solution. However, their success depends on robust net maneuver and closing decisions, which are degraded by uncertainties from noisy sensor observations and imperfect simulations. A new study by Feng Liu, Achira Boonrath, Eleonora M. Botta, and Souma Chowdhury, presented at the 2025 AIAA Aviation Forum, focuses on the first uncertainty source—noisy observations—and propagates its effect through a capture performance metric called the Capture Quality Index (CQI).
The researchers applied two uncertainty quantification (UQ) techniques: Sobol's variance-based sensitivity analysis and a perturbation-based method. They evaluated both a fixed baseline control policy and a neuro-control policy trained to guide the net during deployment. To demonstrate practical trade-offs, they used a high-fidelity simulator alongside a lower-fidelity surrogate-based environment. Results show that sensor noise significantly degrades CQI, and the neuro-control policy offers some robustness but still requires careful UQ. The work provides a framework to design more reliable debris capture missions by accounting for observation errors.
- Study uses Sobol' variance-based sensitivity analysis and perturbation-based methods to quantify uncertainty in tether-net debris capture.
- Evaluates both fixed baseline control and a trained neuro-control policy for net maneuver guidance.
- High-fidelity vs. surrogate-based simulation reveals trade-offs between prediction accuracy and computational ease for UQ.
Why It Matters
Improving uncertainty quantification in tether-net systems boosts reliability of autonomous debris removal, crucial for safe orbital operations.