This paper introduces OrbitalGym, a JAX-based simulation environment designed to study adversarial encounters between satellites. The environment allows for large-scale parallel simulations on GPUs, enabling researchers to determine when planning algorithms or system design have a greater impact on the outcome of orbital games. The study compares various planning methods, including model predictive path integral control and Monte Carlo tree search, across different scenarios like pursuit-evasion and lady-bandit-guard, analyzing factors such as propellant ratio, sensor capability, and game duration. AI
IMPACT Provides a new simulation framework for studying AI planning in complex, adversarial environments.
RANK_REASON The item is a research paper detailing a new simulation environment and its application to a specific problem domain. [lever_c_demoted from research: ic=1 ai=1.0]
- alphaXiv
- CatalyzeX
- DagsHub
- Gotit.pub
- graphics processing unit
- Hugging Face
- Influence Flower
- Jax
- linear-quadratic regulator
- model predictive path integral control
- Monte Carlo tree search
- OrbitalGym
- ScienceCast
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →