FlyCNS: Connectome-Grounded Information Organization for Communication-Constrained Embodied Control

2026-09-23

AI Summary

This study addresses the challenge of organizing information for local computation and whole-body coordination in embodied control under communication constraints. We propose a distributed control framework inspired by the *Drosophila* neural connectome, which leverages the topology of biological connectomes as a weak prior for communication resource allocation. By integrating reinforcement learning, the framework preserves local sensorimotor computation while enabling selective long-range communication, effectively guiding robots to balance computational and communication resources within limited bandwidth. Experimental results demonstrate that the proposed approach maintains high tracking accuracy even under extremely low communication budgets, exhibiting significantly slower performance degradation compared to full-communication baselines. This work provides an efficient bio-inspired solution for multi-agent coordination in resource-constrained scenarios.

Abstract

Robotic bodies are inherently distributed in sensing and actuation, yet learning-based control still commonly relies on centralized information processing. This work studies the problem of information organization in communication-constrained embodied control: which computations should remain local, and which information is worth transmitting for whole-body coordination. We propose FlyCNS, an embodied information-organization framework inspired by the Drosophila brain--nerve-cord connectome. FlyCNS preserves local sensorimotor computation within each limb and enables selective long-range communication through separate ascending and descending routing pathways. From a real connectome, FlyCNS extracts the directional structural complexity of these two pathway types and uses it as a weak prior over communication allocation, while message content, transmission timing, and locomotion policies remain task-adaptive and are learned through reinforcement learning. In Unitree Go1 simulation, FlyCNS exhibits more graceful performance degradation as the communication budget is tightened. Under the most restrictive setting, it uses only about 21--22\% of the communication of the full-communication reference, while still maintaining a tracking score of approximately 0.882 under both command protocols, with a gap of no more than 6.1\% from the full-communication reference. These results indicate that real neural connectomes can inform not only the structural design of control networks, but also provide transferable inductive biases for information organization across embodiments, guiding robots in balancing local computation and long-range coordination under limited communication resources.

Problem

• embodied control
• communication constraints
• information organization
• connectome

Innovation

• Connectome-inspired architecture
• Communication-constrained control
• Embodied information organization
• Reinforcement learning
• Distributed sensorimotor computation