An Embodied Multiagent Framework Based on Token Communications for Cooperative ISAC

2026-10-08

AI Summary

This study addresses the prohibitive signaling overhead and coordination latency caused by direct state sharing in UAV cooperative sensing and communication. To this end, we propose SI-TokCom, a novel framework that pioneers a token-based embodied multi-agent collaboration paradigm. Specifically, it leverages a pretrained vector-quantized codebook to compress exchanged states and intentions, while integrating multi-agent reinforcement learning to jointly optimize token selection and physical control, thereby minimizing energy consumption under rate constraints. Experimental results demonstrate that the proposed approach enables efficient integrated sensing and communication cooperation over low-bandwidth channels, achieving performance closely approaching centralized baselines. Compared with local baselines, it significantly improves both communication and sensing rates while maintaining comparable energy expenditure.

Abstract

The emerging low-altitude economy demands unmanned aerial vehicle (UAV)-enabled integrated sensing and communication (ISAC) for reliable connectivity and environmental awareness. In particular, embodied UAV agents offer a promising means of supporting autonomous operations through a closed loop linking perception, decision-making, and physical actions. However, each UAV has access only to local observations, and effective cooperation requires exchanging local states and intentions. Directly sharing such information can incur substantial signaling overhead and hinder timely coordination in dynamic environments. To deal with this problem, this paper investigates a cooperative ISAC network of embodied UAV agents and formulates a joint token communications (TokCom) and physical control problem to minimize total propulsion energy subject to communication and sensing rate requirements. Then, we propose a state--intent TokCom (SI-TokCom) framework driven by multi-agent embodied policy learning. Specifically, separate pretrained codebooks enable compact exchanges of local states and intentions, while UAV agents jointly learn to select and compose tokens and determine physical actions based on local observations and received tokens. Simulation results show that SI-TokCom achieves 98.9\% and 99.1\% of the centralized baseline's communication and sensing rates, respectively. Compared with the local baseline, it improves the corresponding rates by 5.0\% and 43.6\%, respectively, with essentially unchanged propulsion energy. These results highlight the potential of TokCom for communication-efficient cooperation among embodied UAV agents in ISAC systems.

Problem

• Integrated Sensing and Communication
• Embodied UAV Agents
• Multi-agent Cooperation
• Token Communications
• Signaling Overhead

Innovation

• Token Communications
• Embodied Multiagent
• Integrated Sensing and Communication
• Multi-agent Policy Learning
• UAV Cooperation