SuperNav: An Agentic Navigation System for Any Task in Any Scene
Summary
SuperNav is a navigation framework that equips a pretrained multimodal LLM with an agent harness and navigation tools—without navigation-specific fine-tuning—allowing the model to interpret requests and specify visual destination points while delegating motion execution. It outperforms four baselines on instance-level, multi-object, and demand-driven tasks and is demonstrated on a real quadruped robot.
View Cached Full Text
Cached at: 10/09/26, 06:27 AM
Paper page - SuperNav: An Agentic Navigation System for Any Task in Any Scene
Source: https://huggingface.co/papers/2610.12126
Abstract
General-purposeservicerobotsneednavigationsystemsthatcanhandlediversehumanrequestsinunfamiliarenvironments,combiningtaskgeneralitywithscenegenerality.Someexistingmethodsfine-tunemultimodallargelanguagemodels(MLLMs)topredictnavigationactions,makingtheirbehaviordependentonthecoverageofnavigationtrainingdataandpotentiallylimitinggeneralizationtonewrequestsandenvironments.OurkeyinsightistolettheMLLMfocusoninterpretingrequests,understandingscenes,andmakingdecisionswhilepreservingitsgeneral-purposecapabilitiesanddelegatingmotionexecutiontonavigationtools.Torealizethisidea,weintroduceSuperNav,whichequipsapretrainedMLLMwithaspecializedagentharnesswithoutnavigation-specificfine-tuningoftheMLLM.OurharnesssupportsthesedecisionswithNavigationSkills,agent-orientedToolsforphysicalinteraction,andtask-progressandcontextmanagement.Aunifiedvisual-pointinterfaceconnectsdecision-makingtomotionbyallowingthemodeltospecifydestinationsdirectlyinimagesandreviseitsdecisionsfromexecutionfeedback.Together,thesecomponentssupportsustainednavigationacrossdifferenttaskrequirementsandenvironments.SuperNavoutperformsfourevaluatedbaselinesoninstance-level,multi-object,anddemand-driventasks.Category-levelevaluationonHM3Danddeploymentonarealquadrupedrobotfurtherdemonstrateitsapplicabilityacrossenvironments.ProjectPage:https://zju3dv.github.io/SuperNav/
View arXiv pageView PDFProject pageGitHub10Add to collection
Get this paper in your agent:
hf papers read 2610\.12126
Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash
Models citing this paper0
No model linking this paper
Cite arxiv.org/abs/2610.12126 in a model README.md to link it from this page.
Datasets citing this paper0
No dataset linking this paper
Cite arxiv.org/abs/2610.12126 in a dataset README.md to link it from this page.
Spaces citing this paper1
Collections including this paper0
No Collection including this paper
Add this paper to acollectionto link it from this page.
Similar Articles
LightNav-0: Eliciting VLM Spatial Intelligence for Generalist Embodied Navigation
LightNav-0 is a compact generalist navigation model that leverages pretrained vision-language models' spatial intelligence to achieve state-of-the-art embodied navigation across diverse tasks and robot embodiments.
Qwen-RobotNav Technical Report: A Scalable Navigation Model Designed for an Agentic Navigation System
Qwen-RobotNav is a scalable navigation model with a parameterized interface enabling dynamic task modes and observation parameters, achieving state-of-the-art performance through multi-task training and zero-shot generalization to real-world robotics.
Scaffolding Foundation Models into Physical-World Agents Pushes the Frontier of Long-Horizon Navigation
NavMCP is an agentic scaffolding framework that integrates vision-language models with navigation foundation models to enable persistent long-horizon physical-world exploration, achieving state-of-the-art results on benchmarks and real-robot tasks.
HarnessVLN: Unifying Training-Free Embodied Navigation through an Agent Harness
HarnessVLN is a zero-shot, training-free framework for embodied navigation that unifies perception, retrieval, grounding, navigation, recovery, and termination through a unified tool interface, achieving state-of-the-art results on benchmarks like R2R and RxR.
NavHarness: Towards Lifelong Embodied Navigation
The paper introduces NavHarness, a training-free embodied navigation harness that integrates memory processing into the navigation loop, enabling agents to carry experience across sessions for lifelong navigation. On GOAT-Bench, it improves success rate by up to 22.6 points over context-only sessions and achieves state-of-the-art results with GPT-6/Astra.