Multi-Agent Robotic Control with Onboard Vision-Language Models

• URL: http://arxiv.org/abs/2607.07403v1
• Date: Wed, 08 Jul 2026 13:37:31 GMT
• Title: Multi-Agent Robotic Control with Onboard Vision-Language Models
• Abstract summary: Vision Language Models (VLMs) and Vision Language Action (VLA) models have shown promise in robotic control.<n>They face significant challenges regarding explainability, generalization, and compute requirements.<n>This paper presents a Multi-Agent System (MAS) architecture that addresses these limitations by deploying specialized agents on onboard hardware.
• Score: 38.78172334790013
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Vision Language Models (VLMs) and Vision Language Action (VLA) models have shown promise in robotic control. Yet, they face significant challenges regarding explainability, generalization, and compute requirements. This paper presents a Multi-Agent System (MAS) architecture that addresses these limitations by deploying specialized agents on onboard hardware - eliminating dependence on external compute. The system controls a multi-purpose autonomous mobile manipulator in a simulated industrial warehouse, fulfilling five task categories: safety inspection, warehouse maintenance, warehouse search, package quality verification, and responding to human requests. Compact VLMs (3-20B parameters) are used throughout, with fine-tuning applied to improve package inspection accuracy. A novel "Megamind" orchestration agent mitigates context retention issues inherent to long-horizon planning with smaller models. The system was validated in a hardware-in-the-loop simulation using an AMD Ryzen(TM) AI mini PC. Results demonstrate that a fully onboard MAS architecture is a viable, cost-efficient alternative to cloud-dependent deployments, with strong potential for real-world transfer. The simulation environment has been released as open source under the Apache 2.0 licence.

Related papers

• Show-Harness: Just a VLM Agent Can Play Robots [53.80887083886227]
  Embodied vision-language models (VLMs) exhibit broad intelligence about the world, yet translating this intelligence into robot control remains challenging.<n>We present Show-Harness, an Embodied Harness that enablesVLMs to "play" robots through a compact semantic interface linking intent to action.<n>We further develop GUMI, which extends the same semantic action space to GUI-based demonstration collection, allowing humans and agents to "play" robots across embodiments without specialized teleoperation hardware.
  arXiv  Detail & Related papers  (2026-09-09T17:53:38Z)
• Decoupling the Declarative from the Procedural in Vision-Language-Action Models [65.22639791239932]
  Policy trained to clone a behavior from object-specific demonstrations must generalize beyond that object.<n>We propose w$2$VLA, a new Vision-Language-Action model with restructured information flow.<n>Unlike popular, state-of-the-art VLAs, we show that our modular approach successfully decouples knowledge representations.
  arXiv  Detail & Related papers  (2026-06-19T14:43:58Z)
• PEACE: A Planner-Executor Agent with Constraint Enforcement for UAVs [4.022353603224729]
  We propose a planner-executor agent for PX4-based drones that decouples high-level mission planning from low-level control.<n>A large language model performs single-pass task planning, while execution is handled through a structured ROS 2 tool-calling interface.<n>Results highlight improved explainability, constraint enforcement, and reduced LLM calls compared to tightly coupled LLM control.
  arXiv  Detail & Related papers  (2026-05-26T10:03:22Z)
• GAMMS: Graph based Adversarial Multiagent Modeling Simulator [15.681127447904322]
  We present GAMMS (Graph based Adrial Multiagent Modeling Simulator), a lightweight yet scalable simulation framework.<n>GAMMS emphasizes five core objectives: scalability, ease of use, integration-first architecture, fast visualization feedback, and real-world grounding.<n>It enables efficient simulation of complex domains such as urban road networks and communication systems.
  arXiv  Detail & Related papers  (2026-02-04T22:38:51Z)
• AgenticLab: A Real-World Robot Agent Platform that Can See, Think, and Act [27.922630781100864]
  We present AgenticLab, a model-agnostic robot agent platform and benchmark for open-world manipulation.<n>We benchmark state-of-the-art VLM-based agents on real-robot tasks in unstructured environments.<n>Our benchmark reveals several failure modes that offline vision-language tests fail to capture.
  arXiv  Detail & Related papers  (2026-02-02T05:30:14Z)
• OWMM-Agent: Open World Mobile Manipulation With Multi-modal Agentic Data Synthesis [70.39500621448383]
  Open-world mobile manipulation task remains a challenge due to the need for generalization to open-ended instructions and environments.<n>We propose a novel multi-modal agent architecture that maintains multi-view scene frames and agent states for decision-making and controls the robot by function calling.<n>We highlight our fine-tuned OWMM-VLM as the first dedicated foundation model for mobile manipulators with global scene understanding, robot state tracking, and multi-modal action generation in a unified model.
  arXiv  Detail & Related papers  (2025-06-04T17:57:44Z)
• General-Purpose Aerial Intelligent Agents Empowered by Large Language Models [9.603293922137965]
  This paper presents the first aerial intelligent agent capable of open-world task execution.<n>Our hardware-software co-designed system addresses two fundamental limitations.<n>The system demonstrates reliable task planning and scene understanding in communication-constrained environments.
  arXiv  Detail & Related papers  (2025-03-11T11:13:58Z)
• DeeR-VLA: Dynamic Inference of Multimodal Large Language Models for Efficient Robot Execution [114.61347672265076]
  Development of MLLMs for real-world robots is challenging due to the typically limited computation and memory capacities available on robotic platforms. We propose a Dynamic Early-Exit Framework for Robotic Vision-Language-Action Model (DeeR) that automatically adjusts the size of the activated MLLM. DeeR demonstrates significant reductions in computational costs of LLM by 5.2-6.5x and GPU memory of LLM by 2-6x without compromising performance.
  arXiv  Detail & Related papers  (2024-11-04T18:26:08Z)
• BEHAVIOR Vision Suite: Customizable Dataset Generation via Simulation [57.40024206484446]
  We introduce the BEHAVIOR Vision Suite (BVS), a set of tools and assets to generate fully customized synthetic data for systematic evaluation of computer vision models. BVS supports a large number of adjustable parameters at the scene level. We showcase three example application scenarios.
  arXiv  Detail & Related papers  (2024-05-15T17:57:56Z)
• RoboLLM: Robotic Vision Tasks Grounded on Multimodal Large Language Models [4.4173427917548524]
  Multimodal Large Language Models (MLLMs) have emerged as novel backbones for various downstream tasks. We introduce the RoboLLM framework, equipped with a BEiT-3 backbone, to address all visual perception tasks in the ARMBench challenge.
  arXiv  Detail & Related papers  (2023-10-16T09:30:45Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.