Reproduce “Performance Comparison of ROS2 Middlewares for Multi-robot Mesh Networks in Planetary Exploration” 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
Loïck Pierre Chovet, Gabriel Manuel Garcia, Abhishek Bera, Antoine Richard, Kazuya Yoshida, Miguel Angel Olivares-Mendez
This research provides an empirical performance comparison of FastRTPS, CycloneDDS, and Zenoh ROS 2 middlewares within a multi-robot mesh network, simulating conditions for planetary exploration. The study quantifies their performance across metrics like reachability, delay, and resource usage, identifying Zenoh as the most stable and resource-efficient option for maintaining connectivity in dynamic, challenging environments.
Key insights
- Zenoh consistently outperformed FastRTPS and CycloneDDS in reachability and message delay, particularly during dynamic network topology changes, offering enhanced stability.
- Zenoh demonstrated significantly lower CPU resource utilization (approximately half) on embedded computers compared to DDS implementations, suggesting improved power efficiency, despite slightly higher RAM usage and longer CPU engagement time.
- Selecting a ROS 2 middleware involves trade-offs: Zenoh excels in connectivity and CPU efficiency for dynamic mesh networks, while DDS implementations might offer lower data overhead or RAM usage for specific applications.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.