AgentComm-Bench is the first benchmark to stress-test cooperative embodied AI under realistic wireless impairments like packet loss and bandwidth collapse.
March 24, 2026
Original Paper
AgentComm-Bench: Stress-Testing Cooperative Embodied AI Under Latency, Packet Loss, and Bandwidth Collapse
arXiv · 2603.20285
The Takeaway
Current multi-agent systems assume perfect communication, but this paper reveals that tasks like navigation fail catastrophically (96% drop) under real-world constraints. It provides a standardized protocol and a new 'staleness-aware' fusion method to build more robust robotic swarms.
From the abstract
Cooperative multi-agent methods for embodied AI are almost universally evaluated under idealized communication: zero latency, no packet loss, and unlimited bandwidth. Real-world deployment on robots with wireless links, autonomous vehicles on congested networks, or drone swarms in contested spectrum offers no such guarantees. We introduce AgentComm-Bench, a benchmark suite and evaluation protocol that systematically stress-tests cooperative embodied AI under six communication impairment dimensio