DARL is a two-stage MARL technique that trains robust evasion policies for a single satellite evader against multiple adversarial pursuers in a partially observable orbital game, outperforming standard optimization planners.
UneVEn: Universal Value Explo- ration for Multi-Agent Reinforcement Learning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2024 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Satellite Chasers: Divergent Adversarial Reinforcement Learning to Engage Intelligent Adversaries on Orbit
DARL is a two-stage MARL technique that trains robust evasion policies for a single satellite evader against multiple adversarial pursuers in a partially observable orbital game, outperforming standard optimization planners.