Federated Single-Agent Robotics: Multi-Robot Coordination Without Intra-Robot Multi-Agent Fragmentation

Xue Qin; Simin Luan; John See; Cong Yang; Zhijun Li

cs.ROApr 13, 2026

Federated Single-Agent Robotics: Multi-Robot Coordination Without Intra-Robot Multi-Agent Fragmentation

Authors: Xue Qin, Simin Luan, John See, Cong Yang, Zhijun Li

Organizations: School of Software, Harbin Institute of Technology, Harbin, China · School of Computer Science and Technology, Harbin Institute of Technology, Harbin, China · School of Mathematical and Computer Sciences, Heriot-Watt University, Malaysia Campus, Malaysia · School of Future Science and Engineering, Soochow University, Suzhou, China

Abstract

As embodied robots move toward fleet-scale operation, multi-robot coordination is becoming a central systems challenge. Existing approaches often treat this as motivation for increasing internal multi-agent decomposition within each robot. We argue for a different principle: multi-robot coordination does not require intra-robot multi-agent fragmentation. Each robot should remain a single embodied agent with its own persistent runtime, local policy scope, capability state, and recovery authority, while coordination emerges through federation across robots at the fleet level. We present Federated Single-Agent Robotics (FSAR), a runtime architecture for multi-robot coordination built on single-agent robot runtimes. Each robot exposes a governed capability surface rather than an internally fragmented agent society. Fleet coordination is achieved through shared capability registries, cross-robot task delegation, policy-aware authority assignment, trust-scoped interaction, and layered recovery protocols. We formalize key coordination relations including authority delegation, inter-robot capability requests, local-versus-fleet recovery boundaries, and hierarchical human supervision, and describe a fleet runtime architecture supporting shared Embodied Capability Module (ECM) discovery, contract-aware cross-robot coordination, and fleet-level governance. We evaluate FSAR on representative multi-robot coordination scenarios against decomposition-heavy baselines. Results show statistically significant gains in governance locality (d=2.91, p<.001 vs. centralized control) and recovery containment (d=4.88, p<.001 vs. decomposition-heavy), while reducing authority conflicts and policy violations across all scenarios. Our results support the view that the path from embodied agents to embodied fleets is better served by federation across coherent robot runtimes than by fragmentation within them.

Figures & tables

Provenance	Constructs
Inherited (Papers 1–5)	ECM definition, capability contracts, per-robot policy scope, single-robot recovery levels, human oversight model ( $H_{i}$ )
Extended to fleet	ECM registry (from local to shared $\Gamma$ ), recovery hierarchy (from local to fleet escalation), policy composition (from single-robot to cross-robot $\Pi$ )
New in Paper 6	Federation layer $\Phi$ , trust-scoped delegation ( $\Delta$ ), inter-robot capability request protocol, authority delegation semantics, fleet-level human supervision ( $H_{F}$ ), coordination invariants I1–I5, architectural invariants A1–A5

Table 1: Novelty provenance of Paper 6 constructs.

Panel of Figure 1 — Figure 1: Decomposition-heavy multi-agent architecture (a) versus federated single-agent architecture (b). In (a), each robot is internally fragmented into multiple fleet-visible agents, producing nested cross-agent coordination with blurred boundaries. In (b), each robot remains a single coherent agent; coordination occurs through a governed federation layer.

Panel of Figure 2 — Figure 2: FSAR system model: three-layer architecture. Bottom: local robot runtimes, each containing an agent core, ECM manager, policy engine, recovery manager, and human interface. Middle: fleet federation layer with shared registry, coordination engine, policy resolver, recovery orchestrator, audit service, and trust manager. Top: fleet-level human oversight.

Panel of Figure 3 — Figure 3: Five-phase lifecycle of an inter-robot capability request. Phases 1–3 are driven by the requesting robot $r_{i}$ ; phases 4–5 are driven by the executing robot $r_{j}$ under its own local runtime.

Panel of Figure 4 — Figure 4: Left: the four-dimensional authority tuple. Right: delegation chain non-transitivity—authority from $r_{i}$ to $r_{j}$ and $r_{j}$ to $r_{k}$ does not imply authority from $r_{i}$ to $r_{k}$ .

Panel of Figure 5 — Figure 5: Layered recovery hierarchy with monotone escalation. Recovery begins locally and expands outward only when the current level’s budget or capacity is exhausted.

Symbol	Meaning
$\mathcal{R}_{i}$	Local runtime of robot $r_{i}$ : $(A_{i},E_{i},P_{i},T_{i},Q_{i},H_{i})$
$A_{i}$	Agent identity of $r_{i}$
$E_{i}$	ECM set (capabilities) of $r_{i}$
$P_{i}$	Local policy scope of $r_{i}$
$T_{i}$	Trust relations of $r_{i}$
$Q_{i}$	Recovery state of $r_{i}$

Table 2: Summary of principal notation.

Panel of Figure 6 — Figure 6: Fleet runtime architecture. Each robot contains a self-sufficient local runtime with agent core, ECM manager, policy engine, and recovery manager, connected to the federation layer through a thin federation interface. The federation layer provides shared registry, coordination engine, policy resolver, recovery orchestrator, audit service, trust manager, and human oversight console.

Panel of Figure 7 — Figure 7: Structure of a capability advertisement record in the shared ECM registry. Fields are grouped by function: identity (version, contract), governance (trust and policy requirements), authority profile, operational state (availability), and physical context (embodiment).

Model Element	Architecture Component
$A_{i}$ (agent identity)	Agent Core
$E_{i}$ (capability set)	Local ECM Manager
$P_{i}$ (policy scope)	Local Policy Engine
$T_{i}$ (trust state)	Trust & Authority Manager
$Q_{i}$ (recovery authority)	Recovery Mgr + Recovery Orch.
$H_{i}$ (local supervision)	Local Human Supervision I/F

Table 3: Mapping from formal model elements to architectural components.

Panel of Figure 8 — Figure 8: Sequence diagram for Workflow 1 (Door Relay). Robot A discovers Robot B’s door-opening capability through the registry, issues a governed request, and Robot B executes under its own local runtime.

Panel of Figure 9 — Figure 9: Workflow 3: layered recovery escalation for Robot D’s grasp degradation. Recovery proceeds through all four levels—local retry, peer query, fleet reassignment evaluation, and human escalation—demonstrating monotone escalation with containment attempt at each level.

Mechanism	WF1	WF2	WF3
Inter-robot request	$\bullet$	via recovery	$\bullet$
Trust evaluation	task	—	task
Authority tuple	standard	—	standard
Policy composition	allow	—	review
Registry discovery	single	no peer	conflict
Layered recovery	—	all 4 levels	—

Table 4: FSAR mechanisms exercised by each workflow.

Feature	FSAR	CFC	DHMA	RoboOS	JADE	ALL.
Single-agent robot	$\bullet$	$\bullet$	—	$\circ$	—	$\bullet$
Centralized coord.	—	$\bullet$	—	$\bullet$	—	—
Intra-robot agents	—	—	$\bullet$	$\circ$	$\bullet$	—
Trust-scoped deleg.	$\bullet$	—	—	—	$\circ$	—
Layered recovery	$\bullet$	—	$\circ$	$\circ$	—	$\bullet$
Policy composition	$\bullet$	$\circ$	$\circ$	—	$\circ$	—

Table 5: Architectural comparison of FSAR, evaluation baselines, and published systems.

\bullet

= present;

\circ

= partial.

Metric	FSAR	CFC	DHMA
Coord. success (%)	93.0 $\pm$ 1.9	94.0 $\pm$ 1.0	84.0 $\pm$ 1.0
Governance locality	0.94 $\pm$ 0.01	0.69 $\pm$ 0.01	0.53 $\pm$ 0.01
Recovery containment	0.89 $\pm$ 0.03	0.51 $\pm$ 0.04	0.80 $\pm$ 0.03
Authority conflicts	1.04 $\pm$ 0.03	3.79 $\pm$ 0.12	8.11 $\pm$ 0.04
Policy violations	0.40 $\pm$ 0.01	1.66 $\pm$ 0.06	3.34 $\pm$ 0.05
Reassign. latency (s)	3.01 $\pm$ 0.49	2.07 $\pm$ 0.28	5.18 $\pm$ 0.52

Table 6: Evaluation results across five scenarios (

n{=}100

runs per architecture). Values show mean

\pm

95% CI. Higher is better for success, governance locality, recovery containment, and audit attributability. Lower is better for authority conflicts, policy violations, reassignment latency, and human interventions.

Panel of Figure 10 — Figure 10: Evaluation results across all eight metrics. For metrics where lower is better (marked † ), bars are inverted so taller always indicates better performance; raw values are shown above each bar. FSAR achieves the best or near-best performance on seven of eight metrics, with CFC showing lower reassignment latency due to its centralized routing.

Configuration	Success	Gov. Loc.	Rec. Cont.	Auth. Conf.	Pol. Viol.	Audit
Full FSAR	0.95	0.94	0.90	1.18	0.46	0.98
$-$ Trust evaluation	0.97	0.94	0.90	1.65	0.46	0.96
$-$ Policy composition	0.96	0.95	0.90	1.14	0.64	0.98
$-$ Layered recovery	0.74	0.93	0.97	1.14	0.46	0.98
$-$ Shared registry	0.46	0.94	0.90	1.23	0.46	0.93

Table 7: Ablation results aggregated across all five evaluation scenarios. Each row removes one FSAR component while keeping the others intact.

Metric	Comparison	$t$	$p$	$d$	Sig.
Coord. success	FSAR vs CFC	$-0.33$	0.7440	0.05	n.s.
	FSAR vs DHMA	2.88	0.0047	0.42	$p{<}.01$
	CFC vs DHMA	3.13	0.0022	0.45	$p{<}.01$
Governance locality	FSAR vs CFC	22.63	${<}.0001$	2.91	$p{<}.001$
	FSAR vs DHMA	42.23	${<}.0001$	4.88	$p{<}.001$
	CFC vs DHMA	14.04	${<}.0001$	2.03	$p{<}.001$

Table 8: Statistical significance of pairwise metric comparisons (paired

t

-test,

n{=}100

per architecture). Effect sizes reported as Cohen’s

d

$n$	Arch.	Success	Gov. Loc.	Rec. Cont.	Auth. Conf.	Audit
4	FSAR	0.93	0.94	0.89	1.0	0.98
	CFC	0.94	0.69	0.51	3.8	0.84
	DHMA	0.84	0.53	0.80	8.1	0.53
8	FSAR	0.91	0.94	0.87	1.2	0.98
	CFC	0.88	0.65	0.47	4.8	0.81
	DHMA	0.79	0.49	0.76	10.3	0.48

Table 9: Scaling experiment across fleet sizes

n\in\{4,8,16\}

Multi-Robot Coordination

Multi-robot coordination studies how teams of robots plan, communicate, allocate tasks, and act together safely and efficiently.

See all 317 papers

Jun 10, 2026

CHORUS: Decentralized Multi-Embodiment Collaboration with One VLA Policy

Multi-robot collaboration allows robots to efficiently take on a wide range of tasks, from moving a couch through a doorway to assembling structures on a construction site. However, achieving such coordination in mobile multi-robot settings remains challenging: centralized methods conditioned on the combined observations of a team scale poorly with team size, and decentralized methods that train one policy per robot often require explicit alignment procedures or information sharing at inference time to overcome partial observability. Our key insight is that the visuomotor priors of pretrained vision-language-action (VLA) models should enable reactive, decentralized collaboration from each robot's local observations alone, without these inference-time assumptions. We propose CHORUS, a framework that adapts a single VLA backbone to control diverse, multi-robot teams. At inference time, each robot runs an independent copy of CHORUS, conditioned only on its own observations and a robot-identifying prompt. In real-world experiments including mobile tape measurement, library book handovers, and laundry basket lifting, CHORUS achieves a 64% point improvement over decentralized, from-scratch models, improves reactivity to teammate behavior by 40% points, and outperforms centralized baselines. Together, these results show that a shared VLA backbone is capable of achieving decentralized multi-robot collaboration, without per-robot policies or inter-robot communication at inference.

Ria Doshi, Tian Gao, Annie Chen +2

Stanford University

Mobile Robotics Cross-Embodiment Robot Learning Multi-Robot Coordination Vision-Language-Action Models

Ask PDF

Mar 7, 2026

SwiftBot: A Decentralized Platform for LLM-Powered Federated Robotic Task Execution

Federated robotic task execution systems require bridging natural language instructions to distributed robot control while efficiently managing computational resources across heterogeneous edge devices without centralized coordination. Existing approaches face three limitations: rigid hand-coded planners requiring extensive domain engineering, centralized coordination that contradicts federated collaboration as robots scale, and static resource allocation failing to share containers across robots when workloads shift dynamically. We present SwiftBot, a federated task execution platform that integrates LLM-based task decomposition with intelligent container orchestration over a DHT overlay, enabling robots to collaboratively execute tasks without centralized control. SwiftBot achieves 94.3% decomposition accuracy across diverse tasks, reduces task startup latency by 1.5-5.4x and average training latency by 1.4-2.5x, and improves tail latency by 1.2-4.7x under high load through federated warm container migration. Evaluation on multimedia tasks validates that co-designing semantic understanding and federated resource management enables both flexibility and efficiency for robotic task control.

YueMing Zhang, Shuai Xu, Zhengxiong Li +3

California State University, Long Beach, USA · Case Western Reserve University, USA · University of Colorado Denver, USA +2

Multi-Robot Coordination Edge Computing Large Language Model-Based Robot Planning

Ask PDF

Jun 26, 2026

When Multi-Robot Systems Meet Agentic AI:Towards Embodied Collective Intelligence

Embodied AI is increasingly becoming agentic, shifting robots from perception--control pipelines towards closed-loop systems that can retrieve context, deliberate during execution, monitor feedback, and refine future behavior. In parallel, robotics research has also moved from single-robot autonomy towards multi-robot systems, driven by the need for wider sensing, distributed action, heterogeneous capabilities, and fault tolerance. As AI agents move from single-agent use towards multi-agent collaboration, robotics faces a parallel challenge: robot teams must move beyond sharing maps, task assignments, and datasets towards sharing the state produced by embodied agent loops. This article explores Embodied Collective Intelligence (ECI), a future multi-robot paradigm in which a robot team accumulates and uses world context, task progress, and skill experience as shared resources. Specifically, we first review how embodied AI is becoming agentic and how multi-robot cooperation has evolved. We then present Embodied Collective Intelligence through Co-Perception, Co-Action, and Co-Evolution. Finally, we use an illustrative navigation study to examine one concrete component of the concept: shared world-memory inheritance. The study shows that a newly added robot can benefit from merged team memory, but it is not intended as a full evaluation of the ECI framework. Taken together, the review and conceptual framework motivate Embodied Collective Intelligence as a direction for embodied multi-agent intelligence, while the case study grounds one measurable part of the concept.

Yuxuan Yan, Yuanyuan Jia, Qianqian Yang

College of Information Science and Electronic Engineering, Zhejiang University, Hangzhou, China

Multi-Robot Systems Multi-Robot Coordination Embodied AI Shared-Memory Multi-Agent Systems

Ask PDF

Abstract

Multi-Robot Coordination

Explore similar work

CHORUS: Decentralized Multi-Embodiment Collaboration with One VLA Policy

SwiftBot: A Decentralized Platform for LLM-Powered Federated Robotic Task Execution

When Multi-Robot Systems Meet Agentic AI:Towards Embodied Collective Intelligence