Distributed Virtual Model Control for Scalable Human-Robot Collaboration in Shared Workspace

Yi Zhang; Omar Faris; Chapa Sirithunge; Kai-Fung Chu; Fumiya Iida; Fulvio Forni

cs.ROFeb 19, 2026

Distributed Virtual Model Control for Scalable Human-Robot Collaboration in Shared Workspace

Authors: Yi Zhang, Omar Faris, Chapa Sirithunge, Kai-Fung Chu, Fumiya Iida, Fulvio Forni

Organizations: Department of Engineering, University of Cambridge, Cambridge CB2 1PZ, United Kingdom

Abstract

We present a decentralized, agent agnostic, and safety-aware control framework for human-robot collaboration based on Virtual Model Control (VMC). In our approach, both humans and robots are embedded in the same virtual-component-shaped workspace, where motion is the result of the interaction with virtual springs and dampers rather than explicit trajectory planning. A decentralized, force-based stall detector identifies deadlocks, which are resolved through negotiation. This reduces the probability of robots getting stuck in the block placement task from up to 61.2% to zero in our experiments. The framework scales without structural changes thanks to the distributed implementation: in experiments we demonstrate safe collaboration with up to two robots and two humans, and in simulation up to four robots, maintaining inter-agent separation at around 20 cm. Results show that the method shapes robot behavior intuitively by adjusting control parameters and achieves deadlock-free operation across team sizes in all tested scenarios.

Figures & tables

Fig. 1 : Human-robot collaboration in a pick and place task.

Fig. 2 : Four avoidance spring profiles. Profile 1 (gray):

f_{max}=-40,\,\sigma=0.09

; Profile 2 (light blue):

f_{max}=-60,\,\sigma=0.09

; Profile 3 (medium blue):

f_{max}=-40,\,\sigma=0.18

; Profile 4 (dark blue):

f_{max}=-60,\,\sigma=0.18

Fig. 3 : Multi-robot simulation. Robot 1 and 2 are for two-robot experiment. Robot 1-3 are for three-robot experiment.

Panel of Figure 4 — Fig. 4 : Robot 1 avoiding Robot 2 (left) and a human (right) with the same VMC avoidance spring. In both cases, Robot 1 retreats as the other agent approaches, then returns.

Spring profile	$d_{\min}$ (cm)	$T_{<S_{p}}$ (s)
Spring profile 1 (baseline)	$21.8\pm 2.0$	$4.6\pm 1.8$
Spring profile 2 (increased $f_{max}$ )	$24.4\pm 4.7$	$2.3\pm 1.1$
Spring profile 3 (increased $\sigma$ )	$24.8\pm 9.1$	$0.9\pm 0.9$
Spring profile 4 (both increased)	$30.0\pm 9.8$	$0.5\pm 0.5$
Spring profile 1 and damper	$19.9\pm 2.9$	$4.1\pm 0.5$

TABLE I : Minimum separation

d_{\min}^{RH}

and violation duration

T_{<S_{p}}

under different avoidance profiles and mechanisms.

Fig. 5 : Two UR5 robots transport cubic blocks (black squares) toward their target cells (dashed arrows). The intersection of trajectories represents the potential conflict region.

Panel of Figure 7 — Fig. 6 : Top: without conflict resolution layer, robots entered a deadlock due to force balance at $t=25$ s and did not recover. Bottom: with the conflict resolution, robots resolved the conflict by prioritizing robot 1 at 10 s. Top-Left: robot trajectories (dashed - desired EE position; solid - EE position). Top-Right: balancing forces at $t=150\,s$ , indicating a stall; $\boldsymbol{f}_{goal}$ goal-reaching force, $\boldsymbol{f}_{avoid}$ cumulative avoidance force, $\boldsymbol{f}=\boldsymbol{f}_{goal}+\boldsymbol{f}_{avoid}$ . Bottom-Left: robot trajectories. Blue - prioritized interval, Red - non-prioritized interval. Bottom-Right: top - stall detected at $t=10.87\,s$ , bottom - stall resolved at $t=10.95\,s$

Panel of Figure 8 — Fig. 7 : Completion time, safety and interaction metrics across scenarios and team sizes.

TABLE II : Control parameters for each experiment.

Explore similar work

CHORUS: Decentralized Multi-Embodiment Collaboration with One VLA Policy
Jun 10, 2026Ria Doshi, Tian Gao, Annie Chen +2Mobile Robotics Cross-Embodiment Robot Learning
Stanford University
Cognition to Control - Multi-Agent Learning for Human-Humanoid Collaborative Transport
Mar 4, 2026Hao Zhang, Yisen Li, Ruize Geng +4Human-Robot Collaboration Multi-Agent Collaboration
University of Texas at Arlington. · Carnegie Mellon University
Learning to Assist: Collaborative VLAs for Implicit Human-Robot Collaboration
Jun 10, 2026Leo Xu, Letian Li, Alex Cuellar +1Long-Horizon Robotic Manipulation Human-Robot Collaboration
University of Wisconsin–Madison · Massachusetts Institute of Technology

Jun 10, 2026J/K move · Enter open · S save

CHORUS: Decentralized Multi-Embodiment Collaboration with One VLA Policy

Ria Doshi, Tian Gao, Annie Chen, Chelsea Finn+1

Stanford University

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

Open paper AskPDF arXiv

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.

Abstract

Explore similar work

CHORUS: Decentralized Multi-Embodiment Collaboration with One VLA Policy

Cognition to Control - Multi-Agent Learning for Human-Humanoid Collaborative Transport

Learning to Assist: Collaborative VLAs for Implicit Human-Robot Collaboration