Skip to main content
Search

# Embodied Navigation

7 related articles

Subscribe to updates
  1. arXiv

    HKU-led team releases Compass 1 navigation model: trained on monocular RGB video, it reaches 79.4% open-world success against 17.8% for the strongest baseline

    A group from the University of Hong Kong, Shanghai Innovation Institute, Archon Robotics and other institutions released Compass 1, an embodied navigation foundation model trained on monocular RGB video that adds sparse keyframes and an explicit history map to a unified understanding-and-generation backbone, letting an agent backtrack through nearly two minutes of visual history. Across zero-shot real-world trials in four scenarios (office, terrace, complex outdoor and pedestrian-dense street) and eight tasks, Compass 1 reached 79.4% overall success versus 17.8% for the strongest baseline, AwareVLN; on persistent history backtracking the baseline scored 0% and Compass 1 scored 58%.

    Shanghai Innovation InstituteUniversity of Hong Kong
    Hai Zhang（First author）
    Compass 1Embodied NavigationHistory Representation MapStructured VQA Supervision Protocol
  2. arXiv

    Robot imagines the corridor before takeoff: video diffusion supplies virtual views, RRT finds its own path in the reconstructed obstacle map

    From one RGB image and a task prompt, a video diffusion model generates virtual camera views that ORB-SLAM3 and aligned dense depth turn into a plannable 3D space; planning-feasible scenes rise from 13/19 at P20 to 15/19 at P81 across a 19-scene study.

    Hajira Amjad（First author）
    HorizonNavGeometry-Guided Depth AlignmentVideo Diffusion ModelEmbodied NavigationRapidly-exploring Random Tree (RRT)
  3. arXiv

    Mine Odyssey benchmarks spatial agents on 180 tasks across 30 real-world Minecraft reconstructions, with GPT-6 Astra at 85.6% success and the strongest open-weight model at 23.9%

    The authors introduce Mine Odyssey, a benchmark built from 30 BuildTheEarth Minecraft reconstructions of real-world locations (20 outdoor, 10 indoor, spanning 20 countries and regions on five continents), with 1,059 annotated waypoints and 180 natural-language tasks requiring ordered visits to multiple destinations, paired with a programmable action interface, independent arrival verification, and sandboxed isolation; across eight frontier models, GPT-6 Astra reaches 85.6% success, Claude Opus 5.5 reaches 73.9%, and the strongest open-weight model, DeepSeek-V4.1-Flash, reaches 23.9%, with indoor success generally lower than outdoor.

    Yuxuan Cao（First author）
    Mine OdysseyAgentic Spatial IntelligenceEmbodied NavigationProgrammable Agent Framework
  4. arXiv

    NavHarness carries maps, search records and house notes into the next conversation, lifting GOAT-Bench s-SR by 18.6 to 34.8 points

    NavHarness is a training-free embodied harness that makes memory processing part of the navigation loop: each task opens a fresh multi-round agentic session that inherits prior experience through maps, task records and house notes, while outcome verification and run-end consolidation decide what later sessions inherit; on GOAT-Bench it improves s-SR over context-only independent sessions by 18.6 points with GPT-6 Astra, 22.6 with Opus 5, 34.8 with GPT-4o and 30.3 with Qwen3.8-27B, and with SLAM-estimated poses Astra reaches 83.7 s-SR with 36.9 e-SR on GOAT-Bench and 85.9 s-SR on IR2R-CE.

    Xunyi Zhao（First author）
    NavHarnessEmbodied NavigationAgentic MemoryOutcome Verification
  5. arXiv

    EvolvingNav predicts where targets go with a time-indexed 4D belief, lifting first-inspection success from 45.33% to 61.32% on EvoWorld-Bench

    The work introduces EvolvingNav, which builds a persistence–relocation belief from timestamped 3D object histories and closes the loop with an event-driven predict–observe–replan filter so an agent can infer where a target is when targets may move before the query or during navigation, and it releases EvoWorld-Bench with 54 scenes and 803,680 tasks, reporting improved navigation success and search efficiency over the evaluated baselines in simulation and real-robot experiments.

    Mingjian Gao（First author）
    EvolvingNavPredictive 4D Belief NavigationEvoWorld-BenchEvolving-World NavigationEmbodied Navigation
  6. arXiv

    Six-domain evaluation of GPT-6 Astra as an embodied policy: navigation leads, hybrid control lifts manipulation success, but direct in-hand control and dense-reference locomotion remain unreliable

    This work systematically evaluates GPT-6 Astra as an embodied policy across six domains—gripper manipulation, dexterous manipulation, mobile manipulation, navigation, locomotion, and humanoid loco-manipulation—comparing direct control with hybrid control that cooperates with learned policies or whole-body controllers, finding leading navigation results (92% on RxR instruction following, 82% on HM3D object search), hybrid success of 48% on RoboDojo, 50% on DexJoCo, and 38.7% on RoboCasa365, while direct in-hand control and dense-reference locomotion remain unreliable, with substantial token use and inference latency recorded.

    Galbot Team（First author）
    GPT-6 AstraRobot ManipulationEmbodied NavigationHumanoid Loco-Manipulation
  7. arXiv

    PanoVLN lifts R2R-CE success rate to 77.3%, 11.9 points above the previous best, using panoramic vision

    PanoVLN performs vision-and-language navigation in continuous environments with a 4B vision-language model and RGB-only panoramic input, combining longer action-sequence supervision, confidence-guided execution, decision-centric training data, and fused panoramic geometric features to reach 77.3% and 78.0% success rate on R2R-CE and RxR-CE Val-Unseen, exceeding the previous best by 11.9 and 8.7 percentage points, and enabling real-world quadruped navigation with fewer pauses.

    Zhen Wang（First author）
    PanoVLNVision-and-Language NavigationConfidence-Guided ExecutionEmbodied NavigationVision-Language Models

You’re all caught up