SAGES: AI translates natural language into spacecraft trajectories

Researchers have developed SAGES (Semantic Autonomous Guidance Engine for Space), a framework that converts natural-language commands into safe spacecraft trajectories. The system addresses the need for reliable real-time trajectory generation in autonomous spacecraft, reducing reliance on extensive expert input like waypoints and constraints. SAGES translates high-level human intent into trajectories that respect nonconvex constraints. Experiments in fault-tolerant proximity operations with continuous-time constraint enforcement and on a free-flying robotic platform showed over 90% semantic-behavioral coherence. The work was published on arXiv (ID: 2512.09111).