Reproduce “Joint Mobility Control and MEC Offloading for Hybrid Satellite-Terrestrial-Network-Enabled Robots” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Peng Wei, Wei Feng, Yanmin Wang, Yunfei Chen, Ning Ge, Cheng-Xiang Wang
Researchers from Tsinghua University and partner institutions propose a joint mobility control and MEC offloading scheme for hybrid satellite-terrestrial-network-enabled robots. Their dual-agent Q-learning approach significantly reduces the average task completion time by approximately 16% compared to conventional offloading, by actively adjusting robot velocity based on network conditions and offloading needs.
Key insights
- Actively controlling robot mobility, specifically velocity, can be jointly optimized with MEC offloading to significantly improve performance in hybrid satellite-terrestrial networks.
- The NP-hard problem of minimizing average task completion time can be effectively addressed by decomposing it via Lyapunov optimization and applying a dual-agent reinforcement learning algorithm.
- A successful scheme balances the trade-offs between minimizing task completion time and ensuring the robot completes its physical journey within a reasonable time.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.