Stanford Researchers Develop AI System to Enable Safer, More Efficient Spacecraft Docking
-
Researchers at Stanford have developed ART (Autonomous Rendezvous Transformer), an AI system to optimize trajectories for automated spacecraft docking. This could improve safety and reduce errors.
-
ART integrates AI methods into traditional trajectory optimization pipelines to rapidly generate high-quality trajectory candidates.
-
ART outperformed other ML models in tests. It uses transformer models that efficiently parse trajectories like they parse language.
-
ART was developed at Stanford's Center for AeroSpace Autonomy Research (CAESAR), which collaborates on autonomous aerospace systems.
-
The next step is further developing and testing ART in CAESAR's realistic experimental environment before potential real-world testing in orbit.