NASA deploys deep RL for spacecraft operations scheduling

By PulseAugur Editorial · [1 sources] · 2026-06-20 17:33

Researchers have developed a deep reinforcement learning framework to optimize operations scheduling for NASA's Carruthers Geocorona Observatory mission. This system uses an "activity blocks" abstraction and dynamic action-masking to manage complex constraints like power and thermal limits, generating feasible schedules that outperform traditional heuristics. The framework was implemented as the default scheduler from the mission's start, demonstrating its reliability for real-time spacecraft operations and its ability to retrain within six hours. AI

IMPACT Demonstrates the practical application and reliability of deep reinforcement learning for complex, real-time operational scheduling in scientific missions.

RANK_REASON Academic paper detailing a novel application of deep reinforcement learning for a specific scientific mission. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

NASA deploys deep RL for spacecraft operations scheduling

COVERAGE [1]

arXiv cs.LG TIER_1 English(EN) · Lara Waldrop · 2026-06-20 17:33

Deep RL for Fast Long-Horizon Operations Scheduling on NASA's Carruthers Geocorona Observatory Mission

Spacecraft operations scheduling is a highly constrained, long-horizon combinatorial optimization problem that traditionally relies on heuristics, constraint programming, or manual planning. We present a scalable deep reinforcement learning framework developed and deployed for NA…

COVERAGE [1]

Deep RL for Fast Long-Horizon Operations Scheduling on NASA's Carruthers Geocorona Observatory Mission

RELATED ENTITIES

RELATED TOPICS