@misc{ahn2022do,
title = {{Do As I Can, Not As I Say: Grounding Language in Robotic Affordances}},
author = {Michael Ahn and Anthony Brohan and Noah Brown and Yevgen Chebotar and Omar Cortes and Byron David and Chelsea Finn and Chuyuan Fu and Keerthana Gopalakrishnan and Karol Hausman and Alex Herzog and Daniel Ho and Jasmine Hsu and Julian Ibarz and Brian Ichter and Alex Irpan and Eric Jang and Rosario Jauregui Ruano and Kyle Jeffrey and Sally Jesmonth and Nikhil J Joshi and Ryan Julian and Dmitry Kalashnikov and Yuheng Kuang and Kuang-Huei Lee and Sergey Levine and Yao Lu and Linda Luu and Carolina Parada and Peter Pastor and Jornell Quiambao and Kanishka Rao and Jarek Rettinghouse and Diego Reyes and Pierre Sermanet and Nicolas Sievers and Clayton Tan and Alexander Toshev and Vincent Vanhoucke and Fei Xia and Ted Xiao and Peng Xu and Sichun Xu and Mengyuan Yan and Andy Zeng},
year = {2022},
eprint = {2204.01691},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2204.01691},
}
Drives robots with a frozen frontier LLM plus in-context 'knowledge packages' on 12 sim benchmarks + 8 real robots; reports beating prior bests on 9 of 12, but weak real-robot results (competence, not yet intelligence).用冻结的前沿大模型加上下文“知识包”直接控制机器人，覆盖 12 个仿真基准和 8 台真机；报告在 12 个基准中的 9 个超过此前最好结果，但真机表现较弱——结论是具备了具身“能力”，尚未达到具身“智能”。
@misc{sensenova2026has,
title = {{Has GPT-6-Astra Achieved Embodied Intelligence?}},
author = {SenseNova and Joint Research Team and Zhongang Cai and Ziang Cao and Tailai Chen and Han Chong Chua and Linjun Dai and Hanming Deng and Hokin Deng and Haiwen Diao and Chenjing Ding and Weichen Fan and Xiao Fu and Zerui Gong and Chenyang Gu and Zexin He and Zhi Hou and Bo Li and Qinyue Li and Quanzhou Li and Xiaowan Li and Zhichao Liao and Dahua Lin and Zilong Lin and Bo Liu and Chunxiao Liu and Pengfei Liu and Tianqi Liu and Ziwei Liu and Chen Change Loy and Lewei Lu and Jiapeng Luo and Yuexiao Ma and Liang Pan and Xingang Pan and Oscar Qian and Ziyi Shang and Yuyao Shao and Dawei Su and Dacheng Tao and Qingyi Tao and Quan Wang and Ruisi Wang and Xiaogang Wang and Ziyi Wang and Beichen Wen and Donghao Wu and Jiaxu Wu and Silei Wu and Xiaoming Wu and others},
year = {2026},
eprint = {2610.22471},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2610.22471},
}
@misc{kohaut2026driving,
title = {{Driving, Fast or Slow? Neuro-Symbolic Guidance for Motion Prediction in Multi-Modal Ground Mobility}},
author = {Simon Kohaut and Felix Divo and Julius Hahnewald and Benedict Flade and Julian Eggert and Kristian Kersting and Devendra Singh Dhami},
year = {2026},
eprint = {2606.15251},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2606.15251},
}
A Transformer navigation policy trained on many robots' driving data that adapts to new goal types and robots with light fine-tuning.在多种机器人的导航数据上训练的 Transformer 导航策略，经少量微调即可适配新的目标形式和新机器人。
@misc{shah2023vint,
title = {{ViNT: A Foundation Model for Visual Navigation}},
author = {Dhruv Shah and Ajay Sridhar and Nitish Dashora and Kyle Stachowicz and Kevin Black and Noriaki Hirose and Sergey Levine},
year = {2023},
eprint = {2306.14846},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2306.14846},
}
Adds simulated humanoid avatars and a human-in-the-loop interface to Habitat, so robots can be trained and evaluated on tasks shared with people, such as social navigation and social rearrangement.为 Habitat 加入仿真人形化身和真人在环交互接口，可在与人协作的任务（如社交导航、社交重排）上训练和评测机器人。
@misc{puig2023habitat,
title = {{Habitat 3.0: A Co-Habitat for Humans, Avatars and Robots}},
author = {Xavier Puig and Eric Undersander and Andrew Szot and Mikael Dallaire Cote and Tsung-Yen Yang and Ruslan Partsey and Ruta Desai and Alexander William Clegg and Michal Hlavac and So Yeon Min and Vladimír Vondruš and Theophile Gervet and Vincent-Pierre Berges and John M. Turner and Oleksandr Maksymets and Zsolt Kira and Mrinal Kalakrishnan and Jitendra Malik and Devendra Singh Chaplot and Unnat Jain and Dhruv Batra and Akshara Rai and Roozbeh Mottaghi},
year = {2023},
eprint = {2310.13724},
archivePrefix = {arXiv},
primaryClass = {cs.HC},
url = {https://arxiv.org/abs/2310.13724},
}
The predecessor of ViNT — one goal-conditioned navigation policy trained on pooled data from several different robots, which transfers to new robots including ones not seen in training.ViNT 的前作：在多种不同机器人汇集的数据上训练一个目标条件导航策略，可迁移到包括训练中未见过的新机器人。
@misc{shah2022gnm,
title = {{GNM: A General Navigation Model to Drive Any Robot}},
author = {Dhruv Shah and Ajay Sridhar and Arjun Bhorkar and Noriaki Hirose and Sergey Levine},
year = {2022},
eprint = {2210.03370},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2210.03370},
}
The paper behind Nav2, the standard ROS 2 navigation stack - behavior-tree orchestration of planners, controllers and recovery behaviors, validated by long marathon runs on real robots.ROS 2 标准导航栈 Nav2 的论文：用行为树编排规划器、控制器和恢复行为，并在真实机器人上长距离“马拉松”测试验证。
@misc{macenski2020marathon,
title = {{The Marathon 2: A Navigation System}},
author = {Steve Macenski and Francisco Martín and Ruffin White and Jonatan Ginés Clavero},
year = {2020},
eprint = {2003.00368},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2003.00368},
}
Open C++ library for real-time metric-semantic visual-inertial SLAM: VIO, robust pose graph, fast 3D mesher and dense semantic reconstruction as modular parts.开源 C++ 实时度量-语义视觉惯性 SLAM 库，模块化包含 VIO、鲁棒位姿图、快速三维网格与稠密语义重建。
@misc{rosinol2019kimera,
title = {{Kimera: an Open-Source Library for Real-Time Metric-Semantic Localization and Mapping}},
author = {Antoni Rosinol and Marcus Abate and Yun Chang and Luca Carlone},
year = {2019},
eprint = {1910.02490},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/1910.02490},
}
@misc{savva2019habitat,
title = {{Habitat: A Platform for Embodied AI Research}},
author = {Manolis Savva and Abhishek Kadian and Oleksandr Maksymets and Yili Zhao and Erik Wijmans and Bhavana Jain and Julian Straub and Jia Liu and Vladlen Koltun and Jitendra Malik and Devi Parikh and Dhruv Batra},
year = {2019},
eprint = {1904.01201},
archivePrefix = {arXiv},
primaryClass = {cs.CV},
url = {https://arxiv.org/abs/1904.01201},
}
Open spatiotemporal embodied foundation models (2B, 8B, 30B-A3B MoE) for egocentric understanding, localization, grounded reasoning and planning, plus post-trained Nav, Plan, VLA and CoP variants.开源时空具身基础模型（2B、8B 与 30B-A3B MoE），覆盖第一视角理解、时空定位、物理落地推理与规划，并提供导航、规划、VLA、CoP 等后训练版本。
@misc{dang2026rynnbrain,
title = {{RynnBrain: Open Embodied Foundation Models}},
author = {Ronghao Dang and Jiayan Guo and Bohan Hou and Sicong Leng and Kehan Li and Xin Li and Jiangpin Liu and Yunxuan Mao and Zhikai Wang and Yuqian Yuan and Minghao Zhu and Xiao Lin and Yang Bai and Qian Jiang and Yaxi Zhao and Minghua Zeng and Junlong Gao and Yuming Jiang and Jun Cen and Siteng Huang and Liuyi Wang and Wenqiao Zhang and Chengju Liu and Jianfei Yang and Shijian Lu and Deli Zhao},
year = {2026},
eprint = {2602.14979},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2602.14979},
}
Xiaomi's compact embodied agent that plans, tracks progress and recovers from failures across manipulation and navigation, distilled from stronger teachers with a rubric-driven self-improvement loop.小米具身智能团队的轻量统一具身智能体：在操作与导航中负责任务理解、规划、进度跟踪和失败恢复，由更强的教师模型多教师在线蒸馏而来，并配有基于评分细则的自我改进数据飞轮。
@misc{team2026omnibrain,
title = {{OmniBrain-0: A Unified Agentic Model for Embodied Intelligence}},
author = {Xiaomi Embodied Intelligence Team and Yuechen Luo and Chong He and Fang Li and Shaoqing Xu and Jingkai Wang and Menglun Su and Jiaxin Liu and Shengyin Jiang and Zhiyi Lai and Long Chen and Longlong Wang and Liang Hu and Haochen Liu and Yuncheng Jiang and Hongwei Xie and Jieming Cui and Hanbing Li and Junjie Huang and Zhixiang Duan and Hexin Zhang and Weixiang Liang and Ying Gong and Yuhan Wang and Yong Pan and Pengna Li and Kangyi Wu and Ken Deng and Kun Ma and Fuxi Wen and Zhi-xin Yang and Hangjun Ye},
year = {2026},
eprint = {2610.36013},
archivePrefix = {arXiv},
primaryClass = {cs.CL},
url = {https://arxiv.org/abs/2610.36013},
}
Tests whether agent architecture search, so far used on text agents, transfers to perceptual embodied agents; AgentCanvas makes perception, memory, planning and action wiring searchable.研究原用于文本智能体的智能体架构搜索（AAS）能否迁移到有感知的具身智能体；提出 AgentCanvas，把感知、记忆、规划与动作模块的组合方式变成可搜索的设计空间。
@misc{zhou2026automating,
title = {{Automating the Design of Embodied Agent Architectures}},
author = {Jian Zhou and Sihao Lin and Jin Li and Shuai Fu and Gengze Zhou and Qi Wu},
year = {2026},
eprint = {2606.30111},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2606.30111},
}
VLA for language-guided person following with mask and box grounding supervision and target-conditioned history compression; the 0.8B model reports 77.9% on EVT-Bench, plus a 126-episode quadruped tracking dataset.面向语言引导行人跟随的 VLA：加入掩码与检测框定位监督，并按目标压缩历史视觉信息；0.8B 模型在 EVT-Bench 上平均成功率 77.9%，另发布 126 段四足机器人跟踪数据。
@misc{deng2026dobottrack,
title = {{DobotTrack: Target-Grounded Vision-Language-Action Learning for Embodied Person Tracking}},
author = {Shengjun Deng and Yinghan Zhang and Chengrui Wei and Liu Liang and Songyuan Liu and Pengkun Zheng and Yajun Zheng and Ying Zhang and Yuan Tian and Xi Li and Teng Wang},
year = {2026},
eprint = {2611.09041},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.09041},
}
An LLM agent writes partial task plans over perception, planning and control tools; classical search completes them and geometric constraints are bound through the tools. Run in closed loop on Rainbow Y1 and Unitree G1.大模型智能体基于感知、规划、控制工具生成部分任务规划，再由经典搜索补全，并通过工具满足几何约束；在 Rainbow Y1 与 Unitree G1 上闭环运行，交替观察与操作。
@misc{mishra2026tampagent,
title = {{TAMPAgent: Agentic Belief-Space Planning for Mobile Manipulation}},
author = {Utkarsh A. Mishra and Alperen Degirmenci and Yongxin Chen and Danfei Xu and Tucker Hermans and Caelan Garrett},
year = {2026},
eprint = {2611.03584},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.03584},
}
Fuses per-object VLM plausibility and geometric occupancy into one 3D voxel belief that decides where to look, go or open next; on a bimanual mobile humanoid it beats geometry-only search and LLM planning.把逐物体的 VLM 合理性评分与几何占据证据融合为统一的三维体素信念，决定下一步看哪、去哪、开哪个柜子；在双臂移动人形上优于纯几何搜索和 LLM 规划。
@misc{agarwal2026knowing,
title = {{Knowing Where to Look: Interactive Object Search via Spatial-Semantic Fusion}},
author = {Aditya Agarwal and Lucy Cai and Kelly Lu and Travis Manderson and Nicholas Roy and Tomás Lozano-Pérez and Leslie Pack Kaelbling},
year = {2026},
eprint = {2611.00499},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.00499},
}
Learns latent actions from action-free video, then aligns them with physical motion in post-training. Comes with NavAnywhere: 17.5M observations from 15 sources across ground robots, people and drones.先从无动作标注的视频中学习潜在动作，再在后训练阶段对齐真实运动。配套数据集 NavAnywhere 含 17.5M 条视觉观测，来自 15 个来源，覆盖地面机器人、人和无人机。
@misc{nie2026opennwm,
title = {{OpenNWM: A Generalist Navigation World Model with Latent Action Pretraining}},
author = {Dujun Nie and Ruiqi Song and Zixuan Wang and Hanqing Wang and Yifan Han and Baiyong Ding and Wenzhao Zheng and Hangbin Wu and Long Chen},
year = {2026},
eprint = {2611.01844},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.01844},
}
Predicts how nearby vehicles respond to each candidate action and feeds those consequences to a typed decision model; reports replayed NGSIM collisions falling from 48.2% to 2.8%.预测每个候选动作下周围车辆的反应，再把后果交给类型化决策模型比较；报告在回放的 NGSIM 片段中碰撞率从 48.2% 降至 2.8%。
@misc{qu2026jev,
title = {{Jev with Foresight: Safe Language-Guided Autonomous Driving via a Language World Model}},
author = {Yansong Qu and Jiancong Chen and Samuel Labi},
year = {2026},
eprint = {2611.05448},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.05448},
}
Finds a VLA at 80% on atomic skills drops to 19% when composed into mobile manipulation, diagnoses why, and builds an agentic system around it.发现 VLA 在原子技能上 80% 成功率、组合成移动操作后仅 19%，分析失败原因并据此构建智能体式系统。
@misc{wang2026cracking,
title = {{Cracking Mobile Manipulation: From Failure Diagnosis to Agentic Robot System}},
author = {Zixuan Wang and Jia-Ning Li and Yinzhou Tang and Yanzhao Yu and Biao Wu and Meng Fang and Xin Li and Yiwu Zhong},
year = {2026},
eprint = {2610.19600},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2610.19600},
}
HiRe-MoMa fine-tunes a frozen mobile-manipulation policy in the real world by learning residual corrections, with only occasional human guidance.HiRe-MoMa 在真实环境中通过学习残差修正来微调冻结的移动操作策略，只需偶尔的人类引导。
@misc{gong2026hierarchical,
title = {{Hierarchical Residual Policy Learning for Real-World Mobile Manipulation with Sparse Human Guidance}},
author = {Zhefei Gong and Hefeifei Jiang and Zehao Wang and Jiachen Li},
year = {2026},
eprint = {2610.20607},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2610.20607},
}
Pretrains on internet egocentric video to learn how the view changes when approaching a described target, then aligns to actions in sim; reports 94.9% success on its HabitatGS-LN benchmark plus a real quadruped.先在互联网第一视角视频上预训练，学习"朝描述的目标走近时画面如何变化"，再用仿真数据对齐到动作；论文报告在自建 HabitatGS-LN 基准上成功率 94.9%，并在真实四足机器人上部署。
@misc{yang2026scalenav,
title = {{ScaleNav: Scalable Internet Video Pretraining for Language-Conditioned Local Navigation}},
author = {Yihui Yang and Yuanhong Yu and Jiawei Wang and Zhongrui Yu and Jia Shi and Kun Zhan and Ruizhen Hu and Xiaowei Zhou and Yu Liu and Sida Peng},
year = {2026},
eprint = {2610.25690},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2610.25690},
}
Simulator with embodied NPCs, dialogue and event-driven task states, plus a skill harness for frozen MLLMs; its 360-episode bench reports general MLLMs can beat embodied ones; task updates and HRI stay hard.带具身 NPC、受限对话和事件驱动任务状态机的物理仿真框架，并提供让冻结多模态大模型调用多级技能的执行框架；基于 30 个室内场景构建 360 回合基准，评测 8 个模型，报告通用模型可胜过具身专用模型，任务变更和人机协作仍是难点。
@misc{xu2026tarsphere,
title = {{TARSphere: A Robot-Oriented Full-Stack Framework for Long-Horizon Embodied Agents in Evolving Physical and Social Environments}},
author = {Yicheng Xu and Zhikai Xu and Ziyuan Chen and Yuhang Dong and Tao Zhang and Zhucun Xue and Yong Liu and Jiangning Zhang},
year = {2026},
eprint = {2610.30043},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2610.30043},
}
200K+ navigation-planning samples annotated from camera trajectories up to instructions; RL post-training inside a video world model, used as a cheap stand-in environment, reportedly helps on unseen scenes.包含 20 万余条导航规划样本的数据集，场景多样，标注从相机运动轨迹到高层语义指令；并把视频世界模型当作低成本交互环境做强化学习后训练，报告称在未见场景上有明显提升。
@misc{feng2026embodiedworld,
title = {{EmbodiedWorld-200K: Toward Embodied World Agents via Embodied-Planning Dataset and Interactive World Models}},
author = {X. Feng and J. Tang and Z. Lin and L. Chen and S. Shao and Q. Lu and K. Huang},
year = {2026},
eprint = {2610.25959},
archivePrefix = {arXiv},
primaryClass = {cs.AI},
url = {https://arxiv.org/abs/2610.25959},
}
Two coupled JEPAs share a latent space: one proposes goal-conditioned latent intentions, the other predicts consequences, closing an intention-action-consequence loop. Evaluated on image-goal navigation in Habitat MP3D.两个耦合的 JEPA 共享隐空间：一个按目标提出隐式“意图”，另一个预测动作后果，形成意图-动作-后果闭环，让世界模型直接参与构造规划；在 Habitat MP3D 图像目标导航上评测。
@misc{zhou2026injepa,
title = {{InJePA: Intention-First Latent Planning with Coupled JEPA World Models for Visual Navigation}},
author = {Siming Zhou and Bowen Ma and Ruiran Du and Pengcheng Zheng and Sheng Zheng and Dongshen Han and Zhijie Li and Wei Dong and Fan Mo and Chaoning Zhang},
year = {2026},
eprint = {2610.23896},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2610.23896},
}
Dual-system video-action model for VLN: a slow system turns history and instruction into a compact latent plan, a fast one turns plan + latest view into local trajectories; no video is generated at inference.面向视觉语言导航的双系统视频-动作模型：慢系统根据历史观测与指令生成紧凑的隐式规划，快系统结合该规划与最新观测输出局部轨迹；推理时不生成未来视频。代码与权重称将在该仓库发布。
@misc{wang2026havana,
title = {{HAVANA: A Hierarchical Video-Action Model for Vision-Language Navigation}},
author = {Dongzhihan Wang and Yi Du and Chen Wang and Xiuxian Li and Liang Xu},
year = {2026},
eprint = {2610.24122},
archivePrefix = {arXiv},
primaryClass = {cs.CV},
url = {https://arxiv.org/abs/2610.24122},
}
Off-route VLN agents keep following their stale history ('behavioral inertia'). MOON adds a deviation judge that clears it and a GRPO-trained recovery policy (<1% params); reports +5.2 / +6.0 points on R2R-CE / RxR-CE.发现 VLN 智能体偏离路线后会被历史记忆“惯性”拖住。MOON 给冻结的导航策略加上偏航判断器（判断何时清空历史）和 GRPO 训练的恢复策略，参数增加不到 1%；报告在 R2R-CE / RxR-CE 上分别提升 5.2 / 6.0 个百分点。
@misc{zhou2026moon,
title = {{MOON: Learning When to Forget and How to Move On in Vision-and-Language Navigation}},
author = {Jingmeng Zhou and Seson Sun and Binyao Xu and Lingming Zhang and Hanshi Wang and Fudong Ge and Weiming Hu and Jin Gao and Zhipeng Zhang},
year = {2026},
eprint = {2610.33546},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2610.33546},
}
Adapts a language-guided navigation policy to a new building from one iPhone scan, compiled into a training environment with synthetic instructed episodes. Reports 85% real success vs 49% for a demo-fine-tuned baseline.只需用 iPhone 扫描一次新建筑：扫描被编译成训练环境并自动合成带语言指令的导航回合，再用无评论家的闭式 Pareto 效用适配策略；报告真机成功率 85%，用人类示教后训练的基线为 49%。
@misc{yu2026one,
title = {{One Scan is Enough: Demonstration-Free Adaptation for Language-Guided Navigation}},
author = {Zhuoyuan Yu and Yuxing Long and Du Wang and Junyang Wang and Hongwei Fan and Hao Dong},
year = {2026},
eprint = {2610.33794},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2610.33794},
}
RepresentNav fuses several vision foundation models into a policy-learned space and predicts the next latent before acting (JEPA-style); reports 83.8% on VLNVerse fine-grained (+20.05 pts), real-robot tests.RepresentNav 用一个小型 Transformer 融合多个视觉基础模型各层特征，构建由导航目标自行塑造的表示空间，并在每步动作前预测下一时刻的潜在表示（JEPA 式）；报告 VLNVerse 细粒度成功率 83.8%（领先 20.05 个百分点），并在真机上验证。
@misc{chen2026see,
title = {{See What I Need, Dream What I Want: A Navigation Foundation Model on a Self-Constructed Representation Space}},
author = {Jiawei Chen and Yinxi Wang and Zihan Yang and Meijun Gao and Yaowei Liu and Shuming Zhang and Zecheng Li and Xiangju Mi and Yuxuan Liu and Runzhong Huang and Tianlin Li and Xun Zhang and Zhengyang Wang and Xu Miao and Tong Yang},
year = {2026},
eprint = {2610.31762},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2610.31762},
}
GEMS keeps a 3D Gaussian Splatting map that is updated only where the scene changes, and an agent renders views on demand to answer when/where questions. Introduces GEMS-Bench (15 scenes, 473 QA pairs).GEMS 维护一张只在场景变化处增量更新的 3DGS 地图，智能体按需渲染新视角来回答时空问题和定位物体；同时发布 GEMS-Bench（3 个真实 + 12 个合成环境，473 个问答对），报告在物体定位和时空问答上优于近期基线。
@misc{chen2026gaussian,
title = {{Gaussian Episodic Memory for Long-Term Spatiotemporal Reasoning}},
author = {Ling-Chia Chen and Abdelrhman Werby and Kaan Aytekin and Kai O. Arras},
year = {2026},
eprint = {2610.33158},
archivePrefix = {arXiv},
primaryClass = {cs.CV},
url = {https://arxiv.org/abs/2610.33158},
}
OSMA-Bench: matched scenes that differ only in obstacle type test whether navigation policies slow down and keep clearance by how risky an obstacle is. VLM and VLA policies recognise the risk but rarely act on it.OSMA-Bench 用仅障碍物类型不同的配对场景，检验导航策略能否按障碍物危险程度调整间距与速度，并提出语义敏感度指数 SSI。结果显示 VLM 与 VLA 策略能识别哪些障碍更危险，却很少把这种判断体现在实际运动中；在 VLM 方案中加入显式参数推断步骤有所改善。
@misc{bae2026benchmarking,
title = {{Benchmarking Obstacle-Conditioned Safety in Household Robot Navigation}},
author = {Hyunwoong Bae and Jisu Han and Jiyeon Choi and Dohoon Kim and Taesup Moon},
year = {2026},
eprint = {2610.34674},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2610.34674},
}
CoRL 2026 Spotlight. Benchmark for visual navigation on urban sidewalks, built on Isaac Sim with procedurally generated and real-world environments.CoRL 2026 Spotlight。城市人行道视觉导航基准，基于 Isaac Sim，包含程序生成与真实世界环境。
@misc{liu2026sidewalkbench,
title = {{SidewalkBench: Benchmarking Visual Navigation on Urban Sidewalks}},
author = {Zhizheng Liu and Honglin He and Vivek Alumootil and Akshat Pandya and Brad Squicciarini and Wayne Wu and Bolei Zhou},
year = {2026},
eprint = {2606.16953},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2606.16953},
}
CoRL 2026 Spotlight. Decides whether to wait or reroute at a blocked route by modeling how long each obstacle class tends to persist.CoRL 2026 Spotlight。通过建模各类障碍物通常停留多久，决定在路线受阻时等待还是改道。
@misc{sahak2026oscar,
title = {{OSCAR: Obstacle Survival Curves for Adaptive Robot Navigation}},
author = {Hshmat Sahak and Aoran Jiao and Nicholas Rhinehart and Tim Barfoot},
year = {2026},
eprint = {2606.00990},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2606.00990},
}
A single onboard VLA for drones: one shared vision-language representation serves semantic understanding, spatial reasoning and closed-loop waypoint control.面向无人机的机载 VLA：同一视觉-语言表征同时支撑语义理解、空间推理与闭环航点控制。
@misc{liu2026openflyv,
title = {{OpenFlyV2: An onboard vision-language-action model for aerial perception and navigation}},
author = {Junli Liu and Zhongrui You and Chenhui Li and Zhen Li and Yuyang Tian and Qizhen Chen and Delin Qu and Haoming Song and Dong Wang and Xuelong Li and Zhigang Wang and Bin Zhao},
year = {2026},
eprint = {2610.39287},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2610.39287},
}
A vision-language model turns web-map directions into one atomic command at a time, a predictor draws the path in the image, and RL tracks it; a wheeled-legged robot completes real missions of up to 1.37 km.视觉语言模型把网络地图导航拆成一条条原子指令，预测器在图像上画出路径，强化学习负责跟踪；轮腿机器人完成最长 1.37 公里的真实任务。
@misc{patel2026waytrace,
title = {{WayTrace: A Map-to-Language-to-Trace Hierarchy for Embodied Urban Navigation}},
author = {Manthan Patel and Jonas Frey and Pascal Roth and Marco Hutter and Cesar Cadena},
year = {2026},
eprint = {2611.05129},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.05129},
}
A compact JEPA-plus-flow-matching quadrotor policy with an unusually candid report: cloning failed all 60 missions, DAgger plateaued, and a runtime shield raised simulated forest success from 21.1% to 52.2%.紧凑的 JEPA 加流匹配四旋翼策略，报告异常坦率：行为克隆 60 次任务全败，DAgger 进入瓶颈，运行时安全层将仿真森林任务成功率从 21.1% 提至 52.2%。
Argues VLN success metrics hide a 'last-3-meter grounding gap': agents stop nearby without facing the target. Adds instance-level metrics and REALM, a refinement module for any backbone.指出 VLN 成功率掩盖了“最后 3 米定位缺口”：智能体停在附近却未对准目标。提出实例级指标和可接入任意主干的修正模块 REALM。
@misc{shi2026from,
title = {{From Region Arrival to Instance-Level Grounding in Vision-and-Language Navigation}},
author = {Xiangyu Shi and Ruoxi Yang and Wei Tao and Jiwen Zhang and Yanyuan Qiao and Qi Wu},
year = {2026},
eprint = {2607.03792},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2607.03792},
}
MEND plans UAV-UGV recharge teamwork with masked diffusion, enforcing operational rules at execution through a logic-based action mask; tested on real maps and physical robot teams.MEND 用掩码扩散规划无人机与地面车协同充电任务，在执行时通过逻辑动作掩码强制满足运行约束；在真实地图和实体机器人队伍上测试。
@misc{deng2026collaborative,
title = {{Collaborative Planning for UAV-UGV Teams under Operational Constraints via Masked Error-bounded Neuro-Symbolic Diffusion}},
author = {Zihao Deng and Qianhuang Li and Maggie Wigness and John Rogers and Donghyun Kim and Hao Zhang},
year = {2026},
eprint = {2610.18850},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2610.18850},
}
Adds simulated odour transport to Habitat-Sim using Fire Dynamics Simulator, giving a smell-and-vision navigation benchmark for finding invisible gas leaks and early-stage fires, with a multimodal RL baseline.用 Fire Dynamics Simulator 为 Habitat-Sim 加入气味传输仿真，构成视觉加嗅觉的导航基准，任务是定位不可见的气体泄漏和早期火源，并提供多模态强化学习基线。
@misc{lee2026odorspaces,
title = {{OdorSpaces: Visual-Olfactory Embodied Navigation in 3D Environments}},
author = {Changyeon Lee and Kelin Yu and Zheng Wei and Shuna Ni and Ruohan Gao},
year = {2026},
eprint = {2611.05738},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.05738},
}
Scores delivery-robot navigation by money: cost, revenue and break-even, using Isaac Sim plus public filings and injury data. Reports that none of seven baselines is economically viable yet.以经济账评估配送机器人导航：在 Isaac Sim 中结合公开财报与伤害数据计算成本、收入与盈亏平衡。报告七种基线方法目前都无法盈利。
@misc{seong2025costnav,
title = {{CostNav: A Navigation Benchmark for Real-World Economic-Cost Evaluation of Physical AI Agents}},
author = {Haebin Seong and Sungmin Kim and Yongjun Cho and Myunchul Joe and Geunwoo Kim and Yubeen Park and Sunhoo Kim and Samwoo Seong and Yoonshik Kim and Suhwan Choi and Jaeyoon Jung and Jiyong Youn and Jinmyung Kwak and Sunghee Ahn and Jaemin Lee and Younggil Do and Seungyeop Yi and Woojin Cheong and Minhyeok Oh and Minchan Kim and Seongjae Kang and Youngjae Yu and Yunsung Lee},
year = {2025},
eprint = {2511.20216},
archivePrefix = {arXiv},
primaryClass = {cs.AI},
url = {https://arxiv.org/abs/2511.20216},
}
@misc{lei2025robomemory,
title = {{RoboMemory: A Brain-inspired Multi-memory Agentic Framework for Interactive Environmental Learning in Physical Embodied Systems}},
author = {Mingcong Lei and Honghao Cai and Yuyuan Yang and Yimou Wu and Jinke Ren and Zezhou Cui and Liangchen Tan and Junkun Hong and Gehan Hu and Shuangyu Zhu and Shaohan Jiang and Ge Wang and Junyuan Tan and Zhenglin Wan and Zheng Li and Zhen Li and Shuguang Cui and Yiming Zhao and Yatong Han},
year = {2025},
eprint = {2508.01415},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2508.01415},
}
Low-level VLN action head that predicts a language-conditioned flow field over the visible sector and rolls it out into trajectories. Adds a continuous-time Habitat benchmark.低层视觉语言导航动作头：在可见扇区上预测语言条件的流场并展开为连续轨迹；并提供连续时间的 Habitat 评测基准。
@misc{liu2026cofl,
title = {{CoFL-S: Spatially Queryable Sector Flow Fields for Local Language-Conditioned Navigation}},
author = {Haokun Liu and Zhaoqi Ma and Yicheng Chen and Wentao Zhang and Masaki Kitagawa and Zicen Xiong and Jinjie Li and Moju Zhao},
year = {2026},
eprint = {2607.02222},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2607.02222},
}
Social navigation where the VLM writes a costmap instead of choosing actions, checks upstream tracking, and a fast safety check guards against latency. Real-world tests.社交导航：让 VLM 生成代价地图而非直接决策，并校验上游跟踪，另加高频安全检查应对延迟；真机验证。
@misc{liao2026g,
title = {{G2-Nav: Grounded and Guarded Vision-Language Costmaps for Robot Social Navigation}},
author = {Yuwen Liao and Yihang Lan and Yizhuo Yang and Ruimeng Liu and Xinhang Xu and Shenghai Yuan and Lihua Xie},
year = {2026},
eprint = {2607.16956},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2607.16956},
}
Navigates cluttered spaces by deciding, with VLM affordance reasoning, whether to swipe, pick up or avoid each blocking object. Zero-shot; reports +27 points over the best baseline on a real mobile manipulator.在杂乱环境中导航：用 VLM 推理每个挡路物体该推开、拿起还是绕开。零样本，报告在真实移动操作机器人上比最强基线高 27 个百分点。
@misc{llontop2026messynav,
title = {{MessyNav: Zero-Shot Navigation via Manipulation in Messy Environments}},
author = {Kenneth Llontop and Cherie Ho and Dylan Zhou and Anuva Banwasi and Priya Sundaresan and Carlota Parés-Morlans and Jeannette Bohg},
year = {2026},
eprint = {2611.07417},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.07417},
}
Persistent memory for mobile manipulators: a 3D scene graph updated with what the robot learns by acting (locked cabinets, object locations). Reports 80% progress over a 25-task, 3-hour run.移动操作机器人的持久记忆：在三维场景图中记录通过交互学到的知识（如柜子上锁、物体位置）。报告在连续 25 个任务、3 小时的仿真中任务进度达 80%。
@misc{banwasi2026messymem,
title = {{MessyMem: Learning-from-Doing Memory for Mobile Manipulation}},
author = {Anuva Banwasi and William Muckelroy and Priya Sundaresan and Linfeng Zhao and Jeannette Bohg and Cherie Ho},
year = {2026},
eprint = {2609.15976},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2609.15976},
}
Safe task planning for embodied agents: keeps a long-term semantic graph of the scene so an LLM risk predictor can catch hazards outside the current view and trigger replanning. IS-Bench + real robot.具身智能体的安全任务规划：持续维护场景的长期语义图记忆，让 LLM 风险预测器发现视野外的隐患并触发重规划。在 IS-Bench 与真机上验证。
@misc{li2026safe,
title = {{Safe Task Planning with Long-Term Graph Memory for Embodied Agents}},
author = {Siyuan Li and Taiyan Lang and Aoqi Yan and Jia Yu and Feifan Liu and Yihan Du and Yu Zheng and Xun Wang and Peng Liu},
year = {2026},
eprint = {2609.08444},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2609.08444},
}
SocialNav-SUB: a VQA benchmark testing whether VLMs understand real social navigation scenes (spatial, temporal, social reasoning). The best VLM still trails a rule-based baseline and human consensus.SocialNav-SUB：检验 VLM 能否理解真实社交导航场景（空间、时序与社会推理）的视觉问答基准。最好的 VLM 仍落后于规则基线和人类共识。
@misc{tsoi2026stars,
title = {{STARS: From Spatiotemporal Dynamics to Social Representations in Human-Robot Interaction}},
author = {Nathan Tsoi and Michael J. Munje and Tejas Oberoi and Rishab Maheshwari and Pengen Zheng and Tanush Chauhan and Peter Stone and Joydeep Biswas},
year = {2026},
eprint = {2609.40245},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2609.40245},
}
Mobile manipulation policy that infers motion intent to reweight perception and decodes base and arm actions with a coupled flow-matching head. Reports +23.6 to +33.0 points over DSPv2 on three tasks.移动操作策略：推断运动意图以动态分配感知注意力，并用显式建模底盘-机械臂交互的流匹配解码器输出动作。报告在三个任务上比 DSPv2 高 23.6–33.0 个百分点。
@misc{liu2026incom,
title = {{InCoM: Intent-Driven Perception and Structured Coordination for Mobile Manipulation}},
author = {Jiahao Liu and Cui Wenbo and Zhongpu Xia and Yongliang Wang and Haoran Li and Dongbin Zhao},
year = {2026},
eprint = {2602.23024},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2602.23024},
}
Proactive assistant robot: a VLM reads a state-machine memory of the user's routine to decide when to help, and a VLA executes. The memory is revised daily by reflection, without retraining.主动协助机器人：VLM 读取以有限状态机表示的用户日常流程记忆，决定何时提供何种帮助，再由 VLA 执行；记忆每天通过反思更新，无需重新训练。
@misc{lee2026agentic,
title = {{An Agentic Framework for Proactive Robots with Procedural Assistance Memory}},
author = {Dohyun Lee and Minho Park and Jeonghoon Park and Dongyoon Hwang and Junha Hyung and Sungwon Hwang and Byungkun Lee and Jaegul Choo},
year = {2026},
eprint = {2610.21211},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2610.21211},
}
Collects VLM-agent data by having teachers act through the agent's own skill interface, including where to look. Reports 71.76% on 170 held-out open-world mobile-manipulation episodes and 38% zero-shot on a real robot.让示教者直接通过智能体自身的技能接口（包括“往哪看”）执行动作来采集 VLM 智能体训练数据。报告在 170 个开放世界移动操作测试集上成功率 71.76%，零样本真机 38%。
@misc{qiu2026look,
title = {{Look Before You Move: Agent-Interface-Aligned Supervision for Open-World Mobile Manipulation}},
author = {Ronghe Qiu and Teli Ma and Jiaming Zhou and Zeying Gong and Yiting He and Qiang Huang and Junwei Liang},
year = {2026},
eprint = {2610.15572},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2610.15572},
}
Flips the problem: instead of adapting the robot, decides where to place a few fiducial markers so a planned trajectory can be executed reliably, using a shielded conditional diffusion model. Sim + hardware.反过来思考：不改机器人，而是决定在环境中哪里放少量视觉标记，使规划好的轨迹能可靠执行；方法为带安全约束采样的条件扩散模型。仿真与真机验证。
@misc{phoompho2026task,
title = {{Task-Aware Environment Augmentation for Reliable Navigation via Shielded Conditional Diffusion}},
author = {Bharawee Phoompho and Gokul Puthumanaillam and Yan Miao and Ruben Hernandez and Tim Bretl and Sayan Mitra and Melkior Ornik},
year = {2026},
eprint = {2606.15154},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2606.15154},
}
Diffusion transformer that treats each robot as a token and generates coordinated multi-robot trajectories in one pass, meeting safety and connectivity constraints; trained across team sizes.扩散 Transformer 把每台机器人表示为一个 token，一次前向生成满足安全与连通约束的多机协同轨迹；可在不同团队规模的数据上混合训练。
@misc{bai2026robots,
title = {{Robots as Tokens: Unified Diffusion Transformer for Coordinated Multi-Robot Trajectory Generation}},
author = {Ruofei Bai and Jie Chen and Yuxin Cai and Jun Li and Wei-Yun Yau and Lihua Xie},
year = {2026},
eprint = {2606.15550},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2606.15550},
}
Calibrates VLM semantic maps with online conformal prediction so each object class gets a safety margin with coverage guarantees as the vocabulary grows. Reports 61.6% fewer semantic collisions on MP3D.用在线共形预测校准 VLM 语义地图，使不同类别物体获得有覆盖率保证的安全距离，且支持词表扩展。报告在 MP3D 上语义碰撞减少 61.6%，四足真机覆盖率 90.73%。
RL post-training for navigation diffusion policies with perturbation-based exploration and group-normalized Q-scores, across embodiments. Reports sim success from 61.20% to 84.28% and real hard cases from 10% to 65%.面向导航扩散策略的强化学习后训练：保留预训练先验的扰动探索 + 组归一化 Q 分重加权匹配，跨多种本体训练。报告仿真成功率从 61.20% 升到 84.28%，真机困难场景从 10% 升到 65%。
@misc{yang2026x,
title = {{X-NavDP: Generalizing Navigation Diffusion Policy to Novel Behavior and Embodiments with Group Q-score Reweighted Matching}},
author = {Tianyu Yang and Yiming Zeng and Wenzhe Cai and Yuqiang Yang and Jiaqi Peng and Hui Cheng and Jiangmiao Pang and Tai Wang},
year = {2026},
eprint = {2607.28560},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2607.28560},
}
Multi-robot social navigation: robots fuse features to forecast occupancy maps with uncertainty, and a constrained multi-agent RL policy avoids regions predicted occupied or uncertain. Unity sim + real robots.多机器人社交导航：多机融合特征预测未来占据栅格及其不确定性，约束多智能体强化学习策略据此避开被预测占据或高度不确定的区域。Unity 仿真与真机验证。
@misc{wang2026coopuq,
title = {{CoopUQ: Robust Multi-Robot Social Navigation via Uncertainty-Aware Cooperative Forecasting}},
author = {Zehao Wang and Jianpeng Yao and Yuping Wang and Yiping Li and Jiaqi Li and Zhixu Li and Kanghoon Lee and Jinfeng Wang and Cong Liu and Zhaowei Tan and Jinkyoo Park and Jiachen Li},
year = {2026},
eprint = {2610.26852},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2610.26852},
}
Uses calibrated barrier risk (CVaR along the MPC rollout) as an early warning to switch to a safer planning mode. Reports 83.3% fewer collisions than a safety-filter-only baseline.以校准后的障碍函数风险（MPC 预测轨迹上的 CVaR）作为预警信号，及时切换到更安全的规划模式。报告碰撞比仅用安全滤波器的基线减少 83.3%。
Learns who owns which objects from everyday observation and keeps a long-term spatial memory refreshed by curiosity-driven exploration, so commands like 'find my bottle' work.从日常观察中学习“哪个物品属于谁”，并通过好奇心驱动的探索刷新长期空间记忆，从而执行“找我的水杯”这类个性化指令。
Uses two world models (social dynamics, scene semantics) only at training time to score candidate actions, distilling that foresight into a single-pass policy. Observers rated the motion more human-like than baselines.仅在训练时用两个世界模型（社交动态、场景语义）为候选动作打分，再把这种前瞻能力蒸馏进单次前向的策略。观察者认为其运动比基线更像人。
@misc{hong2026seeing,
title = {{Seeing Beyond the Next Step: World-Model-Guided Human-Like Navigation in Multi-Agent Scenes}},
author = {Seong-Eun Hong and JuYeong Hwang and Jinhyun Kim and Hanyoung Jang and HyeongYeop Kang},
year = {2026},
eprint = {2611.07730},
archivePrefix = {arXiv},
primaryClass = {cs.GR},
url = {https://arxiv.org/abs/2611.07730},
}
Joint task and motion planning for up to nine robots among dense obstacles: waypoints parameterize motion, and a curriculum with modified RLVR passes motion-planner feasibility back to the task planner (BoxNet3D-OBS).面向密集障碍中多达九台机器人的任务与运动联合规划：用航点参数化运动轨迹，并以课程式训练和改进的 RLVR 将运动规划器的可行性反馈传回任务规划器（BoxNet3D-OBS 基准）。
@misc{ji2026navigating,
title = {{Navigating the Clutter: Waypoint-Based Bi-Level Planning for Multi-Robot Systems}},
author = {Jiabao Ji and Yongchao Chen and Yang Zhang and Ramana Rao Kompella and Chuchu Fan and Gaowen Liu and Shiyu Chang},
year = {2026},
eprint = {2604.21138},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2604.21138},
}
Trains a privileged-state expert, freezes its latent interface and action head, then fits a light adapter mapping frozen-VLM features of image + instruction into it: supervised alignment, not end-to-end RL.先用特权状态训练专家策略并冻结其潜空间接口与动作头，再训练轻量适配器把冻结 VLM 提取的图像与指令特征映射到该潜空间：用监督式对齐代替端到端策略优化。
@misc{subedi2026lcla,
title = {{LCLA: Language-Conditioned Latent Alignment for Vision-Language Navigation}},
author = {Nitesh Subedi and Adam Haroon and Samuel Tetteh and Prajwal Koirala and Cody Fleming and Soumik Sarkar},
year = {2026},
eprint = {2602.07629},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2602.07629},
}
Decouples spatial estimation from MLLM planning via an interactive metric world map plus counterfactual reasoning. Reports zero-shot 48.8% SR on R2R-CE and 42.2% on RxR-CE; runs on a TurtleBot 4 and a drone.通过可交互的度量世界表示把空间状态估计与多模态大模型的语义规划解耦，并引入反事实推理。报告零样本 R2R-CE 成功率 48.8%、RxR-CE 42.2%，并在 TurtleBot 4 与自制无人机上实机部署。
@misc{li2026one,
title = {{One Agent to Guide Them All: Empowering MLLMs for Vision-and-Language Navigation via Explicit World Representation}},
author = {Zerui Li and Hongpei Zheng and Fangguo Zhao and Aidan Chan and Jian Zhou and Sihao Lin and Shijie Li and Qi Wu},
year = {2026},
eprint = {2602.15400},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2602.15400},
}
An image generation model reads a language command and paints the goal and traversability masks onto a bird's-eye-view image; cross-view localization curbs drift. A quadruped finishes a 285 m outdoor route.用图像生成模型理解语言指令，在鸟瞰图上标出目标与可通行区域，并以跨视角定位抑制里程计漂移。四足机器人借此完成 285 米的户外长距离导航。
@misc{wang2026pathpainter,
title = {{PathPainter: Transferring the Generalization Ability of Image Generation Models to Embodied Navigation}},
author = {Yijin Wang and Yuru Tian and Xijie Huang and Weiqi Gai and Mo Zhu and Xin Zhou and Yuze Wu and Fei Gao},
year = {2026},
eprint = {2605.07496},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2605.07496},
}
A large foundation model on a remote workstation gives high-level guidance while a light onboard Edge Adapter refines actions at high rate. Reports 40% higher success than baselines with delays up to 6 s.大模型在远端工作站提供高层指引，机载轻量 Edge Adapter 高频修正动作，两路异步运行并端到端微调。在通信延迟最高 6 秒的真实导航中，成功率比基线高 40%。
@misc{hirose2026asyncvla,
title = {{AsyncVLA: An Asynchronous VLA for Fast and Robust Navigation on the Edge}},
author = {Noriaki Hirose and Catherine Glossop and Dhruv Shah and Sergey Levine},
year = {2026},
eprint = {2602.13476},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2602.13476},
}
FlowPilot: a mapless, monocular-RGB sidewalk navigation policy pre-trained with anchored flow matching on fleet data, then tuned on a small set of human interventions by preference learning.FlowPilot：仅用单目 RGB、无需地图的人行道导航策略，先用锚定流匹配在大规模机器人车队数据上预训练，再用少量人工接管数据做偏好学习微调。
@misc{he2026from,
title = {{From Imitation to Alignment: Human-Preference Flow Policies for Long-Horizon Sidewalk Navigation}},
author = {Honglin He and Zhizheng Liu and Yukai Ma and Bolei Zhou},
year = {2026},
eprint = {2606.12603},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2606.12603},
}
Visual-navigation policies say where to go but not how far. PathLift reads sensor depth (RGB-D, stereo or LiDAR) and fits per-waypoint magnitude and lateral residuals, turning unitless paths metric in one pass.视觉导航策略只给方向、不给距离。PathLift 读取传感器深度（RGB-D、双目或激光雷达），为每个航点拟合长度与横向残差，一次前向就把无单位轨迹变为度量路径。
Labels which frontier explains a human demonstrator's next moves, then trains a VLM to choose among BEV-memory candidates. Candidate-level interface transfers zero-shot to wheeled, quadruped and humanoid robots.通过前瞻人类示教，标注最能解释其后续搜索方向的前沿，再训练 VLM 在 BEV 记忆中的候选点之间做选择。这种候选级接口零样本迁移到轮式、四足和人形机器人。
@misc{cai2026intentnav,
title = {{IntentNav: Learning Spatial-Visual Object Navigation from Human Demonstrations}},
author = {Yuxin Cai and Zongtai Li and Maonan Wang and Muyi Bao and Haokun Zhu and Ruofei Bai and Ding Zhao and Zirui Li and Wenshan Wang and Wei-Yun Yau and Ji Zhang and Chen Lv},
year = {2026},
eprint = {2606.08029},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2606.08029},
}
One diffusion transformer jointly generates long-horizon actions and short-horizon latent visual foresight, avoiding multi-step rollouts. Supports image-goal, point-goal and other goal types in one model.用同一个扩散 Transformer 同时生成长时域动作和短时域潜空间视觉预测，避免多步自回归推演；一个模型支持图像目标、点目标等多种目标类型。
@misc{yang2026wam,
title = {{WAM-Nav: Asymmetric Latent World-Action Modeling for Unified Visual Navigation}},
author = {Ning Yang and Yan Huang and Kaiwen Peng and Ziheng He and Kai Wang and Cui Miao and Kailin Lyu and Guo Li and Xiaofeng Wang and Zheng Zhu and Jing Liu and Nianfeng Liu},
year = {2026},
eprint = {2606.04907},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2606.04907},
}
Navigation where goals change mid-episode without notice: memory is split into global/local spatial and long/short temporal beliefs, with goal-conditioned rewards. Tested on a Spot quadruped.研究回合中目标会无提示地改变的导航：把记忆分解为全局/局部空间与长/短时序信念，并配合目标条件奖励。在 Spot 四足机器人上实机验证。
@misc{liang2026glst,
title = {{GLST: Global-Local Spatio-Temporal Memory Decomposition Framework for Online Chained-Goal Navigation}},
author = {Yousef Liang and Ryan Ardalan and Yarden Goraly and Chenhao Zhu and Yue Yin and Jonas Frey and Claire Tomlin and Marco Pavone and Shankar Sastry},
year = {2026},
eprint = {2610.17927},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2610.17927},
}
Mines a Signal Temporal Logic constraint from demos (e.g. social-navigation comfort) by evolutionary search to shield policies, and runs an evolutionary PDDL planner where a local LLM proposes and repairs plans.通过进化搜索从示教数据中挖掘信号时序逻辑约束（如社交导航舒适度）来为策略加安全护盾，并用进化式 PDDL 规划器让本地部署的大模型提出和修复计划。
@misc{nukapotula2026evoplan,
title = {{EvoPlan: Evolutionary Neuro-Symbolic Robot Planning with Spatio-Temporal Guarantees}},
author = {Bhavya Sai Nukapotula and Samin Moosavi and Haoze Wang and Luke Duncan and Diya Shakkottai and Varun Murali and Srinivas Shakkottai},
year = {2026},
eprint = {2607.06724},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2607.06724},
}
UAV target following that conditions a flow-based trajectory generator on predicted target futures and repairs risky trajectory prefixes inside the sampling loop. Best safety-tracking trade-off; Future-MPC is smoother.无人机目标跟随：用预测的目标未来轨迹作为条件引导流式轨迹生成，并在采样循环中修复高风险的可执行前段。安全与跟踪的权衡最好，但 Future-MPC 更平滑。
One VLM acts as teacher (sees the whole trajectory, infers the goal and reasoning) and student (onboard view only), so language-conditioned navigation reasoning is learned with no human labels. 13 real tasks.同一个 VLM 既当教师（看到完整轨迹，推断导航目标与推理过程）又当学生（只看机载观测），无需人工标注即可学会以语言为条件的导航推理。在 13 个真实任务上验证。
@misc{feng2026training,
title = {{Training Generalist Navigation Policies via Action Self-Distillation}},
author = {Bowen Feng and Keqin Wang and Yulong Yang and Amogh Joshi and Christine Allen-Blanchette and Dhruv Shah and Felix Heide},
year = {2026},
eprint = {2610.17593},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2610.17593},
}
Visuomotor quadrotor policy flies through moving narrow gaps, trained in one stage from scratch by differentiable simulation; a signed-distance-gated objective turns the sparse pass-through event into a dense loss.视觉运动四旋翼策略穿越移动的狭窄缝隙，借助可微仿真从零单阶段训练；以到时变缝隙平面的有符号距离门控各项损失，把稀疏的穿越事件变成稠密目标。
RL policies modulate a reference trajectory for compact quadrotor flips with reversible thrust, one policy per direction. In JAX simulation: 32% lower position RMSE and 57% shorter settling than baselines.强化学习策略调制参考轨迹，实现可反向推力四旋翼的紧凑翻转，正翻与回翻各训一个策略。JAX 仿真中位置 RMSE 降低 32%，稳定时间缩短 57%。
@misc{rodriguez2026acrorl,
title = {{AcroRL: Learning Aggressive Quadrotor Inversion using Bidirectional Thrust}},
author = {Gabriel Rodriguez and Henri Sayag and Abhishek Rathod and John Stecklein and Siddharth Saha and Christopher Barngrover and Wennie Tabib},
year = {2026},
eprint = {2605.24301},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2605.24301},
}
Event-camera drone flight: learn the representation offline from a large dataset, then fine-tune the policy online on cheap state only, with no event rendering. Training drops from 52.44 h to 1.86 h.事件相机无人机飞行：先在大规模离线数据上学习表示，再只用轻量状态信息在线微调策略，无需在训练中渲染事件。训练时间从 52.44 小时降到 1.86 小时。
@misc{messikommer2026approximate,
title = {{Approximate Imitation Learning for Event-based Quadrotor Flight in Cluttered Environments}},
author = {Nico Messikommer and Jiaxu Xing and Leonard Bauersfeld and Marco Cannici and Elie Aljalbout and Davide Scaramuzza},
year = {2026},
eprint = {2603.07578},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2603.07578},
}
Wraps VLM action scores for object-goal search in a conformal prediction layer with a user-set risk level, plus collision-free primitives and backtracking. Reports gains on HM3D/MP3D and on a small real robot.为目标物体搜索中的 VLM 动作打分加上共形预测校准层（风险水平可设定），结合无碰撞运动基元与回溯探索。在 HM3D/MP3D 及小型真实机器人上报告提升。
Outdoor long-range navigation where a VLM first reasons on context and only pulls in geometry and motion cues when an interaction stays unclear. Reports real-world success up from 61.8% to 94.2% over MOSU.户外长距离导航：VLM 先基于场景上下文推理，只有交互情形仍不明确时才调取几何与运动信息。真实实验中导航成功率相对前作 MOSU 从 61.8% 提升到 94.2%。
@misc{liang2026mosu,
title = {{MOSU2: Long-range Navigation System with Multi-modal Scene Understanding}},
author = {Jing Liang and Zhenyu Hou and Senthurbavan Kirubaharan and Md Safwan Mondal and Ashok Gudivada and Xuesu Xiao and Dinesh Manocha},
year = {2026},
eprint = {2611.11148},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.11148},
}
Neural inertial odometry across carriers: a router composes carrier-specific experts, and a light residual branch adapts to new ones. Reports 40% lower ATE than learned baselines.面向多种载体的神经惯性里程计：路由器组合各载体专家，新载体通过轻量残差分支适配。报告 ATE 比学习型基线降低 40%。
@misc{zou2026mosaicimu,
title = {{MosaicIMU: Composing Carrier Experts for Generalizable Neural Inertial Odometry}},
author = {Junye Zou and Huiyi Yan and Xinning Xu and Xiaolei Li and Pengkun Zhou and Jinhui Zhang and Ziyang Meng},
year = {2026},
eprint = {2606.09355},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2606.09355},
}
Generates sim-ready rooms by sampling furniture layouts with diffusion priors, then compiling them into collision-free 3D scenes; nav policies trained there beat ones trained on real scans on held-out real scans.先用扩散先验在功能分区内采样家具布局，再“编译”为无碰撞、可直接仿真的 3D 场景；论文报告在其场景中训练的物体导航策略，在留出的真实扫描场景上优于用等量真实扫描训练的策略。
@misc{liu2026scenebloom,
title = {{SceneBloom: Agentic Indoor Environment Generation with Realistic and Diverse Layouts for Embodied Learning}},
author = {Yuhan Liu and Lianhui Qin and Shenji Wan},
year = {2026},
eprint = {2610.32978},
archivePrefix = {arXiv},
primaryClass = {cs.CV},
url = {https://arxiv.org/abs/2610.32978},
}
@misc{castillaarquillo2026mare,
title = {{MARE: An Open-Source Lunar Rover Simulator with Shadow Ground Truth for Perception and Autonomy}},
author = {Raúl Castilla-Arquillo and Miguel Olivares-Mendez},
year = {2026},
eprint = {2610.28465},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2610.28465},
}
Packages VLA policies, perception models and know-how as 250 reusable "embodied skills" and measures how much they help LLM agents on physical tasks.将 VLA 策略、感知模型与经验知识封装成 250 个可复用的“具身技能”，评测它们对大模型智能体完成物理任务的帮助。
@misc{han2026embodied,
title = {{Embodied Skill Bench: Benchmarking How Well Agents with Embodied Skills Perform on Physical Tasks}},
author = {Zhi Han and Zihan Guo and Chenxi Zeng and Zicai Cui and Yue Xin and Ming Zhou and Dongzhan Zhou and Weinan Zhang and Yang Li},
year = {2026},
eprint = {2610.16238},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2610.16238},
}
Sweeps one measured factor per scenario and certifies whether an agent should proceed or stop; finds the top learned CARLA agent at 0.60 decision accuracy against 0.74 for the rule-based expert.每类场景只改变一个可测因素，并认证智能体此时该通过还是该停下；CARLA 中得分最高的学习型智能体决策准确率为 0.60，规则专家为 0.74。
Mobile-manipulation benchmark with 70 task scenarios combining navigation and manipulation in continuous spaces. Finds leading VLMs still struggle with spatial localization.移动操作评测基准：70 个在连续空间中结合导航与操作的任务场景。发现主流 VLM 在空间定位上仍有明显困难。
@misc{li2026rmmbench,
title = {{RMMBench: A Comprehensive Benchmark for Robotic Mobile Manipulation}},
author = {Huapeng Li and Fuxiang Feng and Jinqiu Fan and Shuo Yang and Fengjiao Chen and Xuezhi Cao and Ran Song and Wei Zhang},
year = {2026},
eprint = {2610.05414},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2610.05414},
}
NavDP trains a navigation diffusion policy in simulation with privileged-information guidance, for sim-to-real use in dynamic open-world scenes.NavDP 在仿真中借助特权信息引导训练导航扩散策略，用于动态开放环境的仿真到真实迁移。
@misc{cai2025navdp,
title = {{NavDP: Learning Sim-to-Real Navigation Diffusion Policy with Privileged Information Guidance}},
author = {Wenzhe Cai and Jiaqi Peng and Yuqiang Yang and Yujian Zhang and Meng Wei and Hanqing Wang and Yilun Chen and Tai Wang and Jiangmiao Pang},
year = {2025},
eprint = {2505.08712},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2505.08712},
}
Has a VLM pick among sparse 'points of interest' (frontiers, likely targets, stairs) tied to waypoints and raw views; planners do the motion. The checkable choices allow RLVR on local VLMs without CoT labels.让视觉语言模型在稀疏的“兴趣点”（探索前沿、疑似目标、楼梯等，每个绑定航点与原始视角）中做选择，底层规划器负责运动。这种可验证的决策使其能用 RLVR 训练本地 VLM，无需人工思维链标注。
@misc{peng2025pigeon,
title = {{PIGEON: VLM-Driven Object Navigation via Points of Interest Selection}},
author = {Cheng Peng and Zhenzhe Zhang and Xiaobao Wei and Yanhao Zhang and Heng Wang and Pengwei Wang and Zhongyuan Wang and Cheng Chi and Shanghang Zhang and Jing Liu},
year = {2025},
eprint = {2511.13207},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2511.13207},
}
For each candidate path, a 3D world model imagines several possible scene layouts behind occlusions; a future-aware value map combines them to pick waypoints. Tested on a Go2 quadruped.对每条候选路径，用 3D 世界模型想象遮挡后面多种可能的场景，再用面向未来的价值地图汇总这些想象来选择航点。在 Go2 四足机器人上测试。
@misc{he2025schr,
title = {{Schrödinger's Navigator: Imagining an Ensemble of Futures for Zero-Shot Object Navigation}},
author = {Yu He and Da Huang and Zhenyang Liu and Zixiao Gu and Qiang Sun and Guangnan Ye and Yanwei Fu and Yu-Gang Jiang},
year = {2025},
eprint = {2512.21201},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2512.21201},
}
Broad embodied-AI survey — robots, simulators, embodied perception, interaction, agents and sim-to-real — with the role of multimodal and world models; backed by the widely used HCPLab paper list.覆盖面很广的具身智能综述：机器人本体、仿真器、具身感知、交互、具身智能体与 sim-to-real，并讨论多模态大模型与世界模型的作用；配套中山大学 HCPLab 维护的常用论文列表。
@misc{liu2024aligning,
title = {{Aligning Cyber Space with Physical World: A Comprehensive Survey on Embodied AI}},
author = {Yang Liu and Weixing Chen and Yongjie Bai and Xiaodan Liang and Guanbin Li and Wen Gao and Liang Lin},
year = {2024},
eprint = {2407.06886},
archivePrefix = {arXiv},
primaryClass = {cs.CV},
url = {https://arxiv.org/abs/2407.06886},
}
ABS pairs an agile RL locomotion policy with a learned reach-avoid value network and a recovery policy that takes over near obstacles, so a quadruped can run fast without crashing.ABS 将敏捷的强化学习运动策略与学习的可达-规避价值网络及恢复策略结合，接近障碍时由恢复策略接管，让四足机器人高速奔跑而不碰撞。
@misc{he2024agile,
title = {{Agile But Safe: Learning Collision-Free High-Speed Legged Locomotion}},
author = {Tairan He and Chong Zhang and Wenli Xiao and Guanqi He and Changliu Liu and Guanya Shi},
year = {2024},
eprint = {2401.17583},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2401.17583},
}
@misc{liu2024ok,
title = {{OK-Robot: What Really Matters in Integrating Open-Knowledge Models for Robotics}},
author = {Peiqi Liu and Yaswanth Orru and Jay Vakil and Chris Paxton and Nur Muhammad Mahi Shafiullah and Lerrel Pinto},
year = {2024},
eprint = {2401.12202},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2401.12202},
}
A video VLM that reads the monocular RGB history plus the instruction and outputs the next navigation action directly, without maps, odometry or depth.基于视频的视觉语言模型：只看单目 RGB 历史帧和语言指令，直接输出下一步导航动作，不依赖地图、里程计或深度。
@misc{zhang2024navid,
title = {{NaVid: Video-based VLM Plans the Next Step for Vision-and-Language Navigation}},
author = {Jiazhao Zhang and Kunyu Wang and Rongtao Xu and Gengze Zhou and Yicong Hong and Xiaomeng Fang and Qi Wu and Zhizheng Zhang and He Wang},
year = {2024},
eprint = {2402.15852},
archivePrefix = {arXiv},
primaryClass = {cs.CV},
url = {https://arxiv.org/abs/2402.15852},
}
Extends NaVid into one video VLA model for several navigation tasks (instruction following, object search, question answering, human following).把 NaVid 扩展为单一视频 VLA 模型，统一处理多种导航任务（指令导航、目标搜索、具身问答、跟随人）。
@misc{zhang2024uni,
title = {{Uni-NaVid: A Video-based Vision-Language-Action Model for Unifying Embodied Navigation Tasks}},
author = {Jiazhao Zhang and Kunyu Wang and Shaoan Wang and Minghan Li and Haoran Liu and Songlin Wei and Zhongyuan Wang and Zhizheng Zhang and He Wang},
year = {2024},
eprint = {2412.06224},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2412.06224},
}
A VLA outputs mid-level navigation commands in language that an RL locomotion policy executes, so legged robots follow natural-language routes; adds Isaac Lab VLN benchmarks.VLA 先用语言输出中层导航指令，再由强化学习运动策略执行，让腿式机器人按自然语言指令导航；并提供 Isaac Lab 上的 VLN 基准。
@misc{cheng2024navila,
title = {{NaVILA: Legged Robot Vision-Language-Action Model for Navigation}},
author = {An-Chieh Cheng and Yandong Ji and Zhaojing Yang and Zaitian Gongye and Xueyan Zou and Jan Kautz and Erdem Bıyık and Hongxu Yin and Sifei Liu and Xiaolong Wang},
year = {2024},
eprint = {2412.04453},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2412.04453},
}
CANVAS lets people guide navigation with abstract cues such as verbal commands or rough sketches, even when they are noisy.CANVAS 让人用语音指令或粗略草图等抽象线索引导导航，即使线索不精确。
@misc{choi2024canvas,
title = {{CANVAS: Commonsense-Aware Navigation System for Intuitive Human-Robot Interaction}},
author = {Suhwan Choi and Yongjun Cho and Minchan Kim and Jaeyoon Jung and Myunchul Joe and Yubeen Park and Minseo Kim and Sungwoong Kim and Sungjae Lee and Hwiseong Park and Jiwan Chung and Youngjae Yu},
year = {2024},
eprint = {2410.01273},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2410.01273},
}
MoMa-LLM grounds LLMs in dynamic scene graphs for long-horizon object search with a mobile manipulator in unexplored spaces.MoMa-LLM 用动态场景图为大模型提供依据，让移动操作机器人在未探索环境中长程搜索物体。
@misc{honerkamp2024language,
title = {{Language-Grounded Dynamic Scene Graphs for Interactive Object Search with Mobile Manipulation}},
author = {Daniel Honerkamp and Martin Büchner and Fabien Despinoy and Tim Welschehold and Abhinav Valada},
year = {2024},
eprint = {2403.08605},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2403.08605},
}
Zero-shot object-goal navigation that scores exploration frontiers by vision-language similarity to the target, with no task-specific training; tested in Habitat and on a Spot robot.零样本目标物体导航：用视觉-语言模型对探索前沿与目标的相关度打分来决定往哪走，无需针对任务训练；在 Habitat 和 Spot 机器人上验证。
@misc{yokoyama2023vlfm,
title = {{VLFM: Vision-Language Frontier Maps for Zero-Shot Semantic Navigation}},
author = {Naoki Yokoyama and Sehoon Ha and Dhruv Batra and Jiuguang Wang and Bernadette Bucher},
year = {2023},
eprint = {2312.03275},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2312.03275},
}
PIRLNav — behavior cloning on human ObjectNav demonstrations, then RL fine-tuning of the same policy; the paper reports 65.0% success on Habitat ObjectNav, 5 points above the previous best.PIRLNav：先在人类 ObjectNav 演示上做行为克隆，再用强化学习微调同一策略；论文报告在 Habitat ObjectNav 上成功率 65.0%，比此前最佳高 5 个百分点。
@misc{ramrakhya2023pirlnav,
title = {{PIRLNav: Pretraining with Imitation and RL Finetuning for ObjectNav}},
author = {Ram Ramrakhya and Dhruv Batra and Erik Wijmans and Abhishek Das},
year = {2023},
eprint = {2301.07302},
archivePrefix = {arXiv},
primaryClass = {cs.LG},
url = {https://arxiv.org/abs/2301.07302},
}
A purely LLM-based VLN agent — GPT models read text descriptions of what the robot sees, its history and possible directions, then reason step by step about where to go; zero-shot on R2R, still behind trained models.纯大语言模型的视觉语言导航智能体：GPT 读取视觉观测的文字描述、历史轨迹和可走方向，逐步显式推理下一步去哪；在 R2R 上零样本运行，但仍落后于专门训练的模型。
@misc{zhou2023navgpt,
title = {{NavGPT: Explicit Reasoning in Vision-and-Language Navigation with Large Language Models}},
author = {Gengze Zhou and Yicong Hong and Qi Wu},
year = {2023},
eprint = {2305.16986},
archivePrefix = {arXiv},
primaryClass = {cs.CV},
url = {https://arxiv.org/abs/2305.16986},
}
For VLN in continuous spaces, builds a topological map online from predicted waypoints, plans on it with a cross-modal transformer and moves with an obstacle-avoiding controller; reports >10%/>20% gains on R2R-CE/RxR-CE.面向连续环境视觉语言导航：根据预测的路点在线构建拓扑地图，用跨模态 Transformer 在图上规划，再由避障控制器执行；报告在 R2R-CE、RxR-CE 上分别比此前最好方法提升 10% 和 20% 以上。
@misc{an2023etpnav,
title = {{ETPNav: Evolving Topological Planning for Vision-Language Navigation in Continuous Environments}},
author = {Dong An and Hanqing Wang and Wenguan Wang and Zun Wang and Yan Huang and Keji He and Liang Wang},
year = {2023},
eprint = {2304.03047},
archivePrefix = {arXiv},
primaryClass = {cs.CV},
url = {https://arxiv.org/abs/2304.03047},
}
Builds an object-level 3D scene graph labelled by vision-language models, which an LLM planner can query in natural language.用视觉语言模型构建带语义标注的物体级三维场景图，大模型规划器可以用自然语言查询。
@misc{gu2023conceptgraphs,
title = {{ConceptGraphs: Open-Vocabulary 3D Scene Graphs for Perception and Planning}},
author = {Qiao Gu and Alihusein Kuwajerwala and Sacha Morin and Krishna Murthy Jatavallabhula and Bipasha Sen and Aditya Agarwal and Corban Rivera and William Paul and Kirsty Ellis and Rama Chellappa and Chuang Gan and Celso Miguel de Melo and Joshua B. Tenenbaum and Antonio Torralba and Florian Shkurti and Liam Paull},
year = {2023},
eprint = {2309.16650},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2309.16650},
}
LLMs write Python that calls perception and control APIs, turning instructions into reactive robot programs.大模型编写调用感知与控制 API 的 Python 代码，把指令变成可响应的机器人程序。
@misc{liang2022code,
title = {{Code as Policies: Language Model Programs for Embodied Control}},
author = {Jacky Liang and Wenlong Huang and Fei Xia and Peng Xu and Karol Hausman and Brian Ichter and Pete Florence and Andy Zeng},
year = {2022},
eprint = {2209.07753},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2209.07753},
}
@misc{huang2022inner,
title = {{Inner Monologue: Embodied Reasoning through Planning with Language Models}},
author = {Wenlong Huang and Fei Xia and Ted Xiao and Harris Chan and Jacky Liang and Pete Florence and Andy Zeng and Jonathan Tompson and Igor Mordatch and Yevgen Chebotar and Pierre Sermanet and Noah Brown and Tomas Jackson and Linda Luu and Sergey Levine and Karol Hausman and Brian Ichter},
year = {2022},
eprint = {2207.05608},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2207.05608},
}
@misc{huang2022visual,
title = {{Visual Language Maps for Robot Navigation}},
author = {Chenguang Huang and Oier Mees and Andy Zeng and Wolfram Burgard},
year = {2022},
eprint = {2210.05714},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2210.05714},
}
Chains an LLM (parses landmarks from the instruction), CLIP (finds them in images) and a visual navigation model to follow route instructions on a real robot without language-annotated robot data.将大语言模型（从指令中解析地标）、CLIP（在图像中找到地标）与视觉导航模型串联，无需带语言标注的机器人数据即可在真实机器人上按路线指令导航。
@misc{shah2022lm,
title = {{LM-Nav: Robotic Navigation with Large Pre-Trained Models of Language, Vision, and Action}},
author = {Dhruv Shah and Blazej Osinski and Brian Ichter and Sergey Levine},
year = {2022},
eprint = {2207.04429},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2207.04429},
}
CLIP on Wheels pairs open-vocabulary localization (CLIP-style models, no fine-tuning) with classical exploration for zero-shot object navigation; the Pasture benchmark adds rare, attribute-described and hidden objects.CoWs on Pasture：CLIP on Wheels 把开放词汇定位（CLIP 等模型，无需微调）与经典探索结合，实现零样本目标物体导航；配套的 Pasture 基准加入了少见物体、按属性描述的物体和被遮挡物体。
@misc{gadre2022cows,
title = {{CoWs on Pasture: Baselines and Benchmarks for Language-Driven Zero-Shot Object Navigation}},
author = {Samir Yitzhak Gadre and Mitchell Wortsman and Gabriel Ilharco and Ludwig Schmidt and Shuran Song},
year = {2022},
eprint = {2203.10421},
archivePrefix = {arXiv},
primaryClass = {cs.CV},
url = {https://arxiv.org/abs/2203.10421},
}
Adds fast rigid-body physics and an interactive apartment dataset (ReplicaCAD) to Habitat, plus a home-rearrangement benchmark for mobile manipulators.为 Habitat 加入快速刚体物理与可交互公寓数据集 ReplicaCAD，并提出面向移动操作机器人的家居物品重排基准。
@misc{szot2021habitat,
title = {{Habitat 2.0: Training Home Assistants to Rearrange their Habitat}},
author = {Andrew Szot and Alex Clegg and Eric Undersander and Erik Wijmans and Yili Zhao and John Turner and Noah Maestre and Mustafa Mukadam and Devendra Chaplot and Oleksandr Maksymets and Aaron Gokaslan and Vladimir Vondrus and Sameer Dharur and Franziska Meier and Wojciech Galuba and Angel Chang and Zsolt Kira and Vladlen Koltun and Jitendra Malik and Manolis Savva and Dhruv Batra},
year = {2021},
eprint = {2106.14405},
archivePrefix = {arXiv},
primaryClass = {cs.LG},
url = {https://arxiv.org/abs/2106.14405},
}
1,000 3D scans of real indoor spaces for Habitat; the main scene set behind ObjectNav and many embodied navigation benchmarks.1000 个真实室内空间的 3D 扫描，供 Habitat 使用；ObjectNav 等众多具身导航基准的主要场景集。
@misc{ramakrishnan2021habitat,
title = {{Habitat-Matterport 3D Dataset (HM3D): 1000 Large-scale 3D Environments for Embodied AI}},
author = {Santhosh K. Ramakrishnan and Aaron Gokaslan and Erik Wijmans and Oleksandr Maksymets and Alex Clegg and John Turner and Eric Undersander and Wojciech Galuba and Andrew Westbury and Angel X. Chang and Manolis Savva and Yili Zhao and Dhruv Batra},
year = {2021},
eprint = {2109.08238},
archivePrefix = {arXiv},
primaryClass = {cs.CV},
url = {https://arxiv.org/abs/2109.08238},
}
The design of Hello Robot's Stretch — a slim, lightweight mobile manipulator with a telescoping arm on a lift, built for homes and used widely in home-robot research.Hello Robot 的 Stretch 设计：在升降柱上装伸缩臂的轻巧移动操作机器人，面向家庭环境，被大量家庭机器人研究采用。
@misc{kemp2021design,
title = {{The Design of Stretch: A Compact, Lightweight Mobile Manipulator for Indoor Human Environments}},
author = {Charles C. Kemp and Aaron Edsinger and Henry M. Clever and Blaine Matulevich},
year = {2021},
eprint = {2109.10892},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2109.10892},
}
VLN-CE moves Room-to-Room instructions from a fixed navigation graph into continuous Habitat scenes with low-level actions; the standard setting for realistic VLN.VLN-CE 把 Room-to-Room 指令从固定导航图搬到 Habitat 连续环境中、使用底层动作执行，是更贴近真实的 VLN 标准设定。
@misc{krantz2020beyond,
title = {{Beyond the Nav-Graph: Vision-and-Language Navigation in Continuous Environments}},
author = {Jacob Krantz and Erik Wijmans and Arjun Majumdar and Dhruv Batra and Stefan Lee},
year = {2020},
eprint = {2004.02857},
archivePrefix = {arXiv},
primaryClass = {cs.CV},
url = {https://arxiv.org/abs/2004.02857},
}
SemExp builds an episodic semantic map and learns a policy that picks long-term goals on it, a modular recipe that became a standard ObjectNav baseline in Habitat.SemExp 构建情景语义地图，并学习在地图上选择长期目标的策略；这一模块化方案成为 Habitat 中 ObjectNav 的常用基线。
Splits exploration into a learned neural SLAM module, a global policy that chooses long-term goals and a local policy that reaches them, which is far more sample-efficient than end-to-end RL.将探索拆为可学习的神经 SLAM 模块、选择长期目标的全局策略和到达目标的局部策略，比端到端强化学习样本效率高得多。
@misc{chaplot2020learning,
title = {{Learning to Explore using Active Neural SLAM}},
author = {Devendra Singh Chaplot and Dhiraj Gandhi and Saurabh Gupta and Abhinav Gupta and Ruslan Salakhutdinov},
year = {2020},
eprint = {2004.05155},
archivePrefix = {arXiv},
primaryClass = {cs.CV},
url = {https://arxiv.org/abs/2004.05155},
}
@misc{anderson2017vision,
title = {{Vision-and-Language Navigation: Interpreting visually-grounded navigation instructions in real environments}},
author = {Peter Anderson and Qi Wu and Damien Teney and Jake Bruce and Mark Johnson and Niko Sünderhauf and Ian Reid and Stephen Gould and Anton van den Hengel},
year = {2017},
eprint = {1711.07280},
archivePrefix = {arXiv},
primaryClass = {cs.CV},
url = {https://arxiv.org/abs/1711.07280},
}
Emanuele Cuzzocrea, Giuseppina Iannotti, Luca Rossini, Arturo Laurenzi et al.
Abstract摘要
Robots are increasingly expected to execute natural-language manipulation tasks in real-world environments. This is particularly relevant in household assistance and logistics, where clutter requires reasoning about both task objectives and execution feasibility. This challenge becomes even more pronounced for bimanual humanoid platforms, where coordinating two manipulators introduces additional interaction and kinematic constraints. We present a zero-shot vision-language framework for execution-aware bimanual manipulation that jointly reasons about task objectives and physical feasibility. …
Xueyang Wang, Zehong Shen, Dingkun Wei, Huaijin Pi et al.
Abstract摘要
We study humanoid navigation across multiple destinations under long-horizon language instructions. This task is challenging because robots must maintain progress across successive goals while continuously producing natural, stable whole-body motion through confined spaces. However, task decomposition in existing methods does not by itself ensure progress throughout execution, while waypoint-based navigation does not guarantee natural, route-faithful whole-body motion. Our key idea is to organize navigation into three complementary roles operating at different time scales: a Brain for task …
@misc{wang2026lhnav,
title = {{LHNav: Long-Horizon Humanoid Navigation via Multirate Coordination and Physics-Grounded Action Data}},
author = {Xueyang Wang and Zehong Shen and Dingkun Wei and Huaijin Pi and Changpeng Wang and Xing Zhu and Hujun Bao and Yujun Shen and others},
year = {2026},
eprint = {2611.12136},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.12136},
}
Hai Zhang, Siqi Liang, Yanjun Guo, Shoufeng Zhang et al.
Abstract摘要
Embodied navigation requires agents to achieve persistent spatial awareness and long-horizon execution stability in open-world environments. However, existing navigation architectures struggle to sustain both capabilities, remaining effective only over a handful of recent frames. We present Compass 1, the first embodied navigation foundation model to unify language understanding and video generation while enabling memory backtracking through nearly 2mins of visual history. Without altering the underlying architecture, we introduce History Representation, which integrates sparse visual …
@misc{zhang2026compass,
title = {{Compass 1: Towards Open-World and Persistent Embodied Navigation from Monocular RGB Videos}},
author = {Hai Zhang and Siqi Liang and Yanjun Guo and Shoufeng Zhang and Yunhao Wang and Hangyu Du and Haoguang Mai and Hongchen Li and others},
year = {2026},
eprint = {2611.12087},
archivePrefix = {arXiv},
primaryClass = {cs.CV},
url = {https://arxiv.org/abs/2611.12087},
}
Arian Hajizadeha, Javad Pourrafia, Negar Shenavara, Sana Bashiria et al.
Abstract摘要
Navigation in dynamic environments requires mobile robots to make progress toward their goals while respecting motion limits and avoiding surrounding obstacles. Learned subgoal policies can guide model predictive control beyond its local prediction horizon, but a geometrically suitable subgoal does not necessarily lead to a successfully solved control problem. This paper proposes OASIS, a hierarchical navigation framework that incorporates recent optimization outcomes into reinforcement learning-based subgoal generation. A high-level policy trained using Soft Actor--Critic generates local …
@misc{hajizadeha2026oasis,
title = {{OASIS: Optimization-Aware Subgoal Inference with Safety Constraints for Autonomous Mobile Robot Navigation}},
author = {Arian Hajizadeha and Javad Pourrafia and Negar Shenavara and Sana Bashiria and Seyed Majid Esmailzadeh},
year = {2026},
eprint = {2611.11996},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.11996},
}
Zero-shot object-goal navigation requires autonomous agents to explore unseen environments and locate target objects without task-specific training, posing challenges in perceptual reliability under observation uncertainty and physically grounded goal termination. Although foundation-model-based approaches improve generalization, they remain vulnerable to unreliable long-range perception, hallucination-induced directional bias, and semantically plausible but geometrically incorrect stopping decisions. To address these limitations, we propose CSG-Nav, a zero-shot navigation framework centered …
@misc{chen2026csg,
title = {{CSG-Nav: Confidence-Aware Semantic-Geometric Fusion for Robust Exploration and Accurate Zero-Shot Object-Goal Navigation}},
author = {Shoubin Chen and Rui Xu and Zehao Wu and Kai Zhang and Tao Huang and Zhou Liu and Qingquan Li},
year = {2026},
eprint = {2611.11978},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.11978},
}
Ziwon Yoon, Seok Hwan Kim, Yiyu Chen, Quan Nguyen et al.
Abstract摘要
Humanoid robots have achieved significant progress in locomotion and manipulation, enabling traversal over rough terrain and richer interaction with the environment, such as moving obstacles. However, most traditional navigation methods assume non-interactive environments and focus only on traversal. While Navigation Among Movable Obstacles (NAMO) approaches relax this assumption, prior efforts have largely focused on flat environments and reactive interaction, rather than predicting object movability and planning ahead. These limitations become critical in outdoor rough-terrain scenarios, …
@misc{yoon2026hint,
title = {{HINT-VLM: Humanoid Interactive Navigation over Rough Terrain with Embodiment-Grounded Vision-Language Model}},
author = {Ziwon Yoon and Seok Hwan Kim and Yiyu Chen and Quan Nguyen and Lu Gan and Ye Zhao},
year = {2026},
eprint = {2611.11665},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.11665},
}
Yunyi Gao, Yixing Shi, Yize Liu, Hai-Duong Tran et al.
Abstract摘要
Quadrotors inspecting buildings and tunnels must navigate unknown clutter with a forward-facing depth camera whose limited field of view constrains scene understanding and collision-avoiding navigation. Recent world-model agents learn navigation through imagined experience (e.g., AirDreamer), but remain data-intensive and often lack explicit geometric memory and awareness of what the camera can observe. We propose ScanDreamer, a map-grounded world model that constructs an online 2D scan-memory map from depth images and vehicle pose. Our world model, actor, and critic are conditioned on this …
@misc{gao2026scandreamer,
title = {{ScanDreamer: A Map-Grounded World Model for Quadrotor Navigation}},
author = {Yunyi Gao and Yixing Shi and Yize Liu and Hai-Duong Tran and Nhat-Hoang P. Nguyen and Shuai Wang and Yi Deng and Shengming Liu and others},
year = {2026},
eprint = {2611.11605},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.11605},
}
Recent studies of embodied agents based on frontier multimodal large language models (MLLMs) develop harness engineering around the MLLM, such as harnesses with post-trained policies and recursive self-improvement loops. Yet whether these additions exceed the native capability of the MLLM remains unexplored. We introduce SimplEvo to explore how far an MLLM can improve under direct control when it retains its execution experience in self-written text memories. It requires no weight updates, post-trained policies, or privileged information. The MLLM writes a coordinate-free task memory once …
@misc{yao2026review,
title = {{Review the Episodes, Refine the Lessons: Simple Memory Evolution for Embodied Agents}},
author = {Yifei Yao and Yude Zou and Xing Gao and Guanbo Jiao and Yukai Wang and Jianming Ma and Weinan Zhang and Chunhua Shen},
year = {2026},
eprint = {2611.11509},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.11509},
}
Robots that travel long distances through public outdoor spaces must do more than avoid obstacles: they must remain on terrain suited to their mobility and decide when to proceed, yield, or detour around pedestrians, cyclists, vehicles, traffic signals, and temporary restrictions. These decisions are difficult because their correctness often depends on how a scene evolves, such as whether an approaching car is slowing or a cyclist will cross the robot's path. Vision-language models (VLMs) can interpret such cues, but they reason over images that lack metric geometry and motion. Supplying this …
@misc{liang2026mosu,
title = {{MOSU2: Long-range Navigation System with Multi-modal Scene Understanding}},
author = {Jing Liang and Zhenyu Hou and Senthurbavan Kirubaharan and Md Safwan Mondal and Ashok Gudivada and Xuesu Xiao and Dinesh Manocha},
year = {2026},
eprint = {2611.11148},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.11148},
}
Jaeyoun Choi, Niklas Schmid, Oswin So, Augustinos Saravanos et al.
Abstract摘要
Disturbances encountered during autonomous navigation generally vary across space, often discontinuously, as obstacles redirect wind and changing terrain alters traction. Tracking controllers typically assume sufficient control authority to mitigate disturbances and maintain safety; the regime where this assumption fails remains underexplored. We present Belief-guided Residual Adaptive Control (BRACE), a hierarchical reinforcement learning (RL) framework for reach-avoid navigation under unknown, bounded, spatially varying disturbances. BRACE infers a disturbance belief from model prediction …
@misc{choi2026brace,
title = {{BRACE: Belief-guided Residual Adaptive Control for Safe Navigation under Unknown Disturbances}},
author = {Jaeyoun Choi and Niklas Schmid and Oswin So and Augustinos Saravanos and Chuchu Fan},
year = {2026},
eprint = {2611.10938},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.10938},
}
Reliable navigation with a single monocular limited field-of-view camera is challenging because the lack of depth measurements and limited spatial awareness complicate collision checking. Existing approaches to tackle these challenges fall into two categories. Model-based planners can incorporate explicit collision-avoidance guidance which leads to better generalization but are often computationally expensive and require manual tuning of the algorithmic parameters. In contrast, end-to-end approaches trained by imitation learning can be flexible but typically encode safety only implicitly and …
Kaiyao Liao, Qingfeng Li, Chen Chen, Qing Sun et al.
Abstract摘要
In real homes, assistive robots must repeatedly locate targets from large furniture to tiny tabletop items, as people move objects between household surfaces. Uniformly instantiating detections at different scales as 3D objects can be brittle for small on-surface items; without explicit object-state validation, stale entries may persist after scene changes. Our key insight is to separate persistent 3D furniture entities from revisable surface-object hypotheses, retaining their visual evidence without requiring independent object geometry. We formalize this design as the Hypothesis Scene Graph …
@misc{liao2026hsgnav,
title = {{HSGNav: Hypothesis Scene Graphs for Multi-Scale Lifelong Navigation under Object Relocation}},
author = {Kaiyao Liao and Qingfeng Li and Chen Chen and Qing Sun and Ningbo Gu and Xiaoze Wu and Haochen Zhao and Ying Li and others},
year = {2026},
eprint = {2611.10547},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.10547},
}
Qizhao Chen, Yuanhong Zeng, Shoh Nishino, Anushri Dixit
Abstract摘要
Directing a robot to locate a target object in unseen environments remains a central challenge in embodied AI. Vision-Language Model (VLM)-based methods provide rich semantic reasoning but often produce uncalibrated and overconfident action decisions. We introduce CPNav, a navigation framework that combines collision-aware planning with statistically calibrated VLM action scoring for map-free object-goal navigation. From RGB-D input, CPNav constructs a navigability map, generates collision-free motion primitives, and queries two VLMs for goal detection and action scoring. The resulting scores …
Zhaofan Zhang, Yue Chang, Rufeng Chen, Hechang Chen et al.
Abstract摘要
Ensuring safe and reliable navigation under diverse sources of uncertainty remains a central challenge for autonomous systems. Within distributional reinforcement learning (DistRL), Implicit Quantile Networks (IQN) have attracted interest for navigation under uncertainty because they enable risk adaptation through flexible quantile sampling. Existing approaches, however, typically use a fixed or randomly sampled Conditional Value-at-Risk (CVaR) level during training and adjust the risk level during deployment using hand-tuned hyperparameters or external sensor signals. This scalar-CVaR …
Shubham Singh, Anand Singh, Anjali Tiwari, Lara Briñón-Arranz et al.
Abstract摘要
Circumnavigating a stationary target with a team of robots requires each robot to hold its own standoff distance, to stay evenly spaced from its neighbours, and to avoid collisions with the target and with each other at every instant. The task becomes arduous when the robots are assigned different orbital radii, as their sensors or roles may demand. This paper presents a safety-critical control framework for this problem. Each robot commands both its acceleration and its turn-rate, measures only its distance to the target, and exchanges information only with its two ring neighbours. The …
We focus on the problem of high-speed multi-robot navigation in cluttered spaces with tight spatial constraints. A key challenge is balancing safety with agility, as existing methods often trade one for the other (slowing down near a bottleneck to avoid collisions versus failing to swerve in time at high speeds), especially for car-like systems. Our approach belongs to the family of formation control methods, in which Ackermann-steering agents coordinate while navigating tight spaces. However, current formation control techniques rely on rigid, pre-defined shapes and transitions that depend …
Sateesh Kumar, Arnav Balaji, Arpit Bahety, Skand Peri et al.
Abstract摘要
Visuomotor policies trained through imitation learning are highly sensitive to changes in camera viewpoint. This is particularly problematic for mobile manipulation (MoMa), where robots rarely execute manipulation from the exact same base pose used during demonstration collection due to navigation uncertainty, localization drift, and environmental constraints. Recent approaches use novel-view synthesis (NVS) to augment demonstrations with unseen viewpoints, but they assume static scenes and therefore fail to re-render the manipulated object during the post-contact interaction phase, leaving …
@misc{kumar2026momasplat,
title = {{MOMASPLAT: Dynamic Novel-View Demonstration Augmentation for Mobile Manipulation}},
author = {Sateesh Kumar and Arnav Balaji and Arpit Bahety and Skand Peri and Georgios Pavlakos and Roberto Mart\'in-Mart\'in},
year = {2026},
eprint = {2611.08372},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.08372},
}
Iaroslav Kolomiets, Iaroslav Sheipak, Mikhail Reppo, Vsevolod Egorov et al.
Abstract摘要
Vision-based rover autonomy requires integrating perception, localization, planning and platform-specific control under limited sensing and computing resources. We present an open-source ROS 2 stack for following directional arrows, reaching a terminal cone and returning autonomously to the start. It connects YOLOv8n detection and geometric direction, range and bearing estimation with RGB-D SLAM, wheel-inertial state estimation, kinematics-aware planning, behavior-tree mission management and STM32 actuation. Standard interfaces separate reusable autonomy from the hardware layer; a …
@misc{kolomiets2026integrated,
title = {{An Integrated Open-Source Stack for Vision-Based Autonomous Navigation of Wheeled Rovers}},
author = {Iaroslav Kolomiets and Iaroslav Sheipak and Mikhail Reppo and Vsevolod Egorov and Viktor Chertopolokhov and Isaac Chairez},
year = {2026},
eprint = {2611.08100},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.08100},
}
Goal interpretation is to translate a natural language instruction into a formal symbolic goal that a planner can execute. It is critical for embodied agents because any error here propagates to every downstream planning and action step. Doing it well requires both generating suitable candidates and reliably selecting among them. We present a two-stage generate-then-rank approach to Goal Interpretation in the Embodied Agent Interface (EAI). First, supervised fine-tuning adapts a Large Language Model (LLM) to EAI's output convention. Second, Candidate-Output Ranking from Errors (CORE) learns …
Seong-Eun Hong, JuYeong Hwang, Jinhyun Kim, Hanyoung Jang et al.
Abstract摘要
Selecting a next motion in a populated scene is not just a matter of avoiding what is currently in the way: an action can be locally feasible yet drift onto the road instead of the crosswalk, cut into a stream of crossing pedestrians, or bottleneck a narrow passage. These failure modes are invisible from the current state alone, and acting on them requires reasoning about what each candidate action will cause. World models offer exactly this, namely action-conditioned predictions of the future, but rolling them out at every decision step is a cost deployable navigation systems cannot pay, and …
@misc{hong2026seeing,
title = {{Seeing Beyond the Next Step: World-Model-Guided Human-Like Navigation in Multi-Agent Scenes}},
author = {Seong-Eun Hong and JuYeong Hwang and Jinhyun Kim and Hanyoung Jang and HyeongYeop Kang},
year = {2026},
eprint = {2611.07730},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.07730},
}
Xuenmin Hu, Qingxue Qian, Wei Luo, Benyi Ma et al.
Abstract摘要
Legged robots equipped with manipulators have shown significant potential in the fields of daily life and industry in complex environments. However, existing methods mainly focus on global navigation or manipulation independently, while overlooking the last-mile navigation problem, where the robot must jointly determine the base placement, approach direction, and grasp execution to achieve reliable object grasping, and different manipulation stages require distinct control behaviors. To address these challenges, we propose Last-Mile Grasp-Navigation (LMGN), a unified policy learning model for …
@misc{hu2026lmgn,
title = {{LMGN: A Unified Policy Learning Model for Last-Mile Grasp-Navigation}},
author = {Xuenmin Hu and Qingxue Qian and Wei Luo and Benyi Ma and Bo Tang and Long Chen},
year = {2026},
eprint = {2611.07515},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.07515},
}
Kenneth Llontop, Cherie Ho, Dylan Zhou, Anuva Banwasi et al.
Abstract摘要
Real-world environments are messy, with paths often blocked by everyday objects like chairs and toys, requiring robots to perform interactive navigation by manipulating obstacles to clear a path. Existing methods rely on extensive simulation-based learning or per-scene instructions specifying which objects to avoid or manipulate. They struggle to generalize across novel objects, diverse environments, and varied contexts such as a cart of fragile glassware that should be avoided while an empty one can be pushed. We formalize this problem as open-world interactive navigation and present …
@misc{llontop2026messynav,
title = {{MessyNav: Zero-Shot Navigation via Manipulation in Messy Environments}},
author = {Kenneth Llontop and Cherie Ho and Dylan Zhou and Anuva Banwasi and Priya Sundaresan and Carlota Par\'es-Morlans and Jeannette Bohg},
year = {2026},
eprint = {2611.07417},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.07417},
}
Pranay Samineni, Adam Thorpe, John Chubick, Anand Balakrishnan et al.
Abstract摘要
Heterogeneous air-ground teams can use aerial sensing to address the limited range and viewpoint of ground-vehicle sensors, but deciding when the UGV needs assistance and what information the UAV should collect remains a challenge. This paper presents a planner-conditioned air-ground assistance method that evaluates how uncertain map information affects the UGV's current MPPI rollouts. When navigation remains degraded, we score the same rollouts under resolved map hypotheses, select the map layer with the largest useful improvement, request the corresponding aerial measurement, fuse the …
@misc{samineni2026when,
title = {{When to Ask for Help: Uncertainty-Triggered Aerial Refinement for Off-Road Navigation}},
author = {Pranay Samineni and Adam Thorpe and John Chubick and Anand Balakrishnan and Christian Ellis and Swarat Chaudhuri and Ufuk Topcu},
year = {2026},
eprint = {2611.06702},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.06702},
}
We present a shared framework that adapts Cosmos world foundation models for real-time interactive world simulation in Physical AI. By streaming visual observations in response to actions and scene controls, the framework enables users and embodied agents to interact with evolving virtual environments. It supports both predicting the visual consequences of actions and generating observations from physical state maintained by an external simulator. We demonstrate this approach across camera navigation and robot manipulation, controllable video transfer, and multi-view autonomous vehicle (AV) …
@misc{nvidia2026interactive,
title = {{Interactive World Simulation with Cosmos for Physical AI}},
author = {NVIDIA and : and Anis Ahmad and Jon Allen and Malik Aqeel Anwar and Jesse Archer and Maciej Bala and Piotr Barejko and others},
year = {2026},
eprint = {2611.06627},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.06627},
}
Sihang Chen, Chen Ye, Wei Chen, Xiaohan Lei et al.
Abstract摘要
In the real world, mobile manipulation robots must perform complex object interactions and make multi-step decisions, often relying on task-relevant information that is no longer observable. However, existing benchmarks provide limited evaluation of such memory-dependent decision-making in mobile manipulation. To address this gap, we introduce RoboCasa-Mem, a new benchmark comprising nine mobile manipulation tasks that require agents to use interaction history at key decision points. Specifically, the tasks are grouped into two categories of memory: One is environment memory, which captures …
@misc{chen2026robocasa,
title = {{Robocasa-Mem: Benchmarking Environment and Process Memory in Long-Horizon Mobile Manipulation}},
author = {Sihang Chen and Chen Ye and Wei Chen and Xiaohan Lei and Zhipeng Tang and Sha Zhang and Yuan Zhang and Zixuan Chen and others},
year = {2026},
eprint = {2611.06475},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.06475},
}
Closed-loop driving benchmarks in CARLA rank agents by a driving score that multiplies route completion by an infraction penalty. This score does not distinguish an agent that stops in front of an obstacle from one that pushes past it. In a corridor closed by two parked cars, for example, one agent that drives through with contact averages 52.2 on the blocked widths, while the agents that stop average between 37.8 and 40.6. We propose failure-boundary profiles, an evaluation in which each scenario family varies one measured factor, such as corridor width or the time gap in oncoming traffic, …
Changyeon Lee, Kelin Yu, Zheng Wei, Shuna Ni et al.
Abstract摘要
Olfaction provides critical information about airborne chemical cues, yet it remains largely underexplored in embodied AI, with limited prior machine olfaction work focusing primarily on semantic odor perception. We introduce OdorSpaces, a physics-based simulation platform and benchmark for visual-olfactory embodied navigation. OdorSpaces augments Habitat-Sim with 3D odor transport and dispersion simulated by Fire Dynamics Simulator (FDS), enabling agents to navigate photorealistic indoor environments using both visual and olfactory observations. The benchmark includes two representative …
@misc{lee2026odorspaces,
title = {{OdorSpaces: Visual-Olfactory Embodied Navigation in 3D Environments}},
author = {Changyeon Lee and Kelin Yu and Zheng Wei and Shuna Ni and Ruohan Gao},
year = {2026},
eprint = {2611.05738},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.05738},
}
Manthan Patel, Jonas Frey, Pascal Roth, Marco Hutter et al.
Abstract摘要
Web services provide walking directions for nearly every city, but following them reliably over long routes remains challenging for autonomous robots. Point-goal navigation policies rely heavily on accurate GNSS, while most vision-language navigation models struggle to execute even short instructions outdoors. We present WayTrace, a hierarchical system that connects web-map directions to robot motion through language. An off-the-shelf orchestrator vision-language model reads the turn-by-turn directions, a satellite map and recent camera images, and issues one atomic language command at a …
@misc{patel2026waytrace,
title = {{WayTrace: A Map-to-Language-to-Trace Hierarchy for Embodied Urban Navigation}},
author = {Manthan Patel and Jonas Frey and Pascal Roth and Marco Hutter and Cesar Cadena},
year = {2026},
eprint = {2611.05129},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.05129},
}
Abhijeet M. Kulkarni, Saimouli Katragadda, Yuze Du, Zijun Sha et al.
Abstract摘要
By harnessing the inherent commonsense capabilities of pretrained Large Language Models (LLMs) and Vision-Language Models (VLMs), current Vision-Language Action(VLA) models for legged robots excel at interpreting language-based navigation commands but lack the physical grounding and explicit safety assurance required for real-world operation. Therefore, this paper aims to cage Legged-Vision-Language Navigation (VLN) through a safety architecture that provides physical grounding and ensures navigation safety. The key idea is not to abandon the powerful capabilities of end-to-end VLA planning, …
@misc{kulkarni2026caging,
title = {{Caging Vision-Language Navigation for Legged Robots via Barrier-Certified Safety Architecture}},
author = {Abhijeet M. Kulkarni and Saimouli Katragadda and Yuze Du and Zijun Sha and Guoquan Huang},
year = {2026},
eprint = {2611.04972},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.04972},
}
Aditya Agarwal, William Shen, Sahit Chintalapudi, Lucy Cai et al.
Abstract摘要
A mobile manipulator asked to "put the toys away in the toy box" must decide which objects are toys and which container the user means, then find base poses, grasps, and placements it can execute. Because a scene representation is built before the instruction is known, it may lack the attributes, affordances, or concepts needed to resolve the task. We present FEIJOA, a language-model agent that grounds the instruction at test time. It queries a grounding tool over a posed RGB-D scan, inspects the returned candidates, and refines its queries until they resolve the instruction. An …
@misc{agarwal2026feijoa,
title = {{FEIJOA: Task-Driven Grounding and Execution for Mobile Manipulation}},
author = {Aditya Agarwal and William Shen and Sahit Chintalapudi and Lucy Cai and Nishanth Kumar and Aidan Curtis and Travis Manderson and Tom\'as Lozano-P\'erez and others},
year = {2026},
eprint = {2611.04080},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.04080},
}
Yang Qu, Qiang Fu, Zonghao Ying, Jiayi Zhao et al.
Abstract摘要
Embodied agents are increasingly deployed in open-world environments, yet current robustness evaluations remain biased toward hazardous instructions and jailbreak vulnerabilities. A more fundamental issue is largely overlooked: agents are rarely tested under semantically sufficient but contextually minimal instructions (i.e., common in real-world human communication), where auxiliary contextual cues are absent. For the first, we identify a novel phenomenon termed Ambient Context Illusion, where agents fail to execute tasks correctly when non-essential contextual details are omitted, despite …
@misc{qu2026vaguebench,
title = {{VagueBench: Benchmarking the Robustness of Embodied Agents on Ambient Context Illusions}},
author = {Yang Qu and Qiang Fu and Zonghao Ying and Jiayi Zhao and Yaoyuan Zhang and Jinyang Guo and Jian Yang and Aishan Liu and others},
year = {2026},
eprint = {2611.03765},
archivePrefix = {arXiv},
primaryClass = {cs.AI},
url = {https://arxiv.org/abs/2611.03765},
}
Tony Tao, Jason Jingzhou Liu, Ruslan Salakhutdinov, Deepak Pathak
Abstract摘要
General-purpose mobile manipulation requires a combination of capable robot hardware, scalable demonstration collection, expressive whole-body control, and effective policy learning. Yet these components are often developed independently, creating mismatches that limit the capabilities of the complete system. We present Pixel, a co-designed full-stack system for learning general-purpose mobile manipulation. Pixel combines a low-cost wheeled humanoid with human-aligned kinematics and compliant grippers, multiple teleoperation and robot-free human data-collection interfaces, a unified …
@misc{tao2026pixel,
title = {{Pixel: A Co-Designed Full-Stack Robot System for Learning General-Purpose Mobile Manipulation}},
author = {Tony Tao and Jason Jingzhou Liu and Ruslan Salakhutdinov and Deepak Pathak},
year = {2026},
eprint = {2611.03648},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.03648},
}
Utkarsh A. Mishra, Alperen Degirmenci, Yongxin Chen, Danfei Xu et al.
Abstract摘要
Open-world robotic mobile manipulation requires acting in physical environments with only partial information. While recent Large Vision-Language Model (LVLM) agents show strong open-world reasoning capabilities, existing architectures that rely on direct tool calling struggle with geometric uncertainty and thus are limited to fully-observable tabletop environments. In contrast, Task and Motion Planning (TAMP) systems can solve long-horizon tasks under partial observability but incur high computational costs and require closed worlds. We propose \textbf{TAMPAgent}, an agentic planning …
@misc{mishra2026tampagent,
title = {{TAMPAgent: Agentic Belief-Space Planning for Mobile Manipulation}},
author = {Utkarsh A. Mishra and Alperen Degirmenci and Yongxin Chen and Danfei Xu and Tucker Hermans and Caelan Garrett},
year = {2026},
eprint = {2611.03584},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.03584},
}
Jakob Wolfram, Pascal Roth, Fan Yang, Cesar Cadena et al.
Abstract摘要
Navigating crowded urban environments requires robots to avoid pedestrians while adhering to social norms such as remaining on sidewalks. Existing approaches typically address either crowd avoidance or semantic path planning in isolation. We propose a reinforcement learning (RL) navigation policy that jointly reasons about semantic terrain and dynamic obstacles from a single semantically annotated bird's-eye-view (BEV) map, without relying on explicit human-state estimation. Our policy outperforms established crowd-navigation and semantic-aware baselines and transfers zero-shot to a legged …
@misc{wolfram2026snap,
title = {{SNAP: Semantic Navigation Among Pedestrians}},
author = {Jakob Wolfram and Pascal Roth and Fan Yang and Cesar Cadena and Marco Hutter},
year = {2026},
eprint = {2611.03498},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.03498},
}
Nikola Fischer, Harry Robertshaw, S. M. Hadi Sadati, Xingyu Chen et al.
Abstract摘要
Mechanical thrombectomy (MT) is an effective treatment for acute ischemic stroke, yet access remains limited by the availability of neurointerventional expertise. Telerobotic and AI-assisted systems may help address this challenge, but it remains unclear how clinicians interact with different levels of autonomy and how autonomous navigation systems perform in realistic environments. This work introduces Copilot Assistance Modes (CAMs), a framework for clinician-AI collaboration in robotic endovascular navigation spanning the autonomy spectrum from manual operation to full autonomy. In the …
@misc{fischer2026clinician,
title = {{Clinician-AI Collaboration in Endovascular Neurointerventions: In-Vitro User Study of Copilot Assistance Modes for Robotic Navigation Across the Autonomy Spectrum}},
author = {Nikola Fischer and Harry Robertshaw and S. M. Hadi Sadati and Xingyu Chen and Weiyuan Deng and Benjamin Jackson and Michael Adofo Kwakye and Sara Sciacca and others},
year = {2026},
eprint = {2611.03446},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.03446},
}
Ball-balancing robots (ballbots) must lean to accelerate, and the lean is limited by the wheel torques and by safety. Planners that use the full robot model are expensive, while planners that ignore the balance dynamics produce references that cannot be followed. This paper presents a two-layer navigation scheme for a three-omni-wheel ballbot. The upper layer is a model predictive controller on a \emph{jerk model} of the ball position, with discrete-time control barrier functions for static, moving, rotating and concave obstacles. The lower layer is a hierarchical sliding mode controller with …
Juho Kalliokoski, Evan G. Center, Steven M. LaValle, Timo Ojala et al.
Abstract摘要
Virtual reality experiments involving navigation and simulated robot motion often require large, realistic environments and numerous distinct trajectories to avoid repetitive navigation while providing sufficient variation for controlled experiments. Designing such scenarios manually is labor-intensive, particularly when trajectories must satisfy predefined geometric and kinematic constraints while remaining diverse and comparable across experimental conditions. This paper presents TRAILS, an automated framework for generating complete virtual navigation scenarios consisting of an environment …
@misc{kalliokoski2026trails,
title = {{TRAILS: Trajectory Generation for Automated Indoor Locomotion Studies}},
author = {Juho Kalliokoski and Evan G. Center and Steven M. LaValle and Timo Ojala and Basak Sakcak},
year = {2026},
eprint = {2611.03254},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.03254},
}
Stefano Trepella, Andrea Ostuni, Mauro Martini, Irene Borgini et al.
Abstract摘要
Social navigation is essential for autonomous service robots operating in human environments, but progress remains constrained by the lack of extensive real-world benchmarks. This paper presents the Real Social Navigation Testbed (RESONATE), a reproducible benchmark for human-aware navigation based on a unified robotic platform, a shared ROS 2 software stack, precise trajectory ground truth, and structured social interaction scenarios. We evaluate eight controller configurations, combining classical and predictive local planners with different social modeling strategies, over eight …
@misc{trepella2026resonate,
title = {{RESONATE: a Real Social Navigation Testbed for Autonomous Service Robots}},
author = {Stefano Trepella and Andrea Ostuni and Mauro Martini and Irene Borgini and Marcello Chiaberge},
year = {2026},
eprint = {2611.02937},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.02937},
}
Shubh Garg, Divyansh Pandey, Vedant Khandelwal, Niyati Rawal et al.
Abstract摘要
Embodied agents that answer questions about their route need a memory of what they saw, and scene graphs are a natural form for it. A vision-language model (VLM) can receive such a graph as text or through a graph encoder, but accuracy alone cannot show whether the graph holds the episode's facts, reaches the model, or is actually relied on. In this diagnostic study, we introduce a graph-swap test for the last of these: each episode is given another episode's graph, and we count how many answers change. On an embodied question-answering benchmark with routes through two to seven rooms, we use …
@misc{garg2026do,
title = {{Do Embodied VLMs Actually Use Their Scene-Graph Memory?}},
author = {Shubh Garg and Divyansh Pandey and Vedant Khandelwal and Niyati Rawal and Amit Sheth},
year = {2026},
eprint = {2611.02056},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.02056},
}
Shubh Garg, Divyansh Pandey, Niyati Rawal, Vedant Khandelwal et al.
Abstract摘要
Embodied question answering asks an agent to navigate, observe, and then answer questions about what it saw, yet answer accuracy alone cannot tell whether any of this happened. A plausible answer can come from the question, the room label, or the answer choices, and a correct answer does not show that the supporting observation was ever made. We present a multi-room benchmark whose 200 test episodes and 800 questions are stratified by route extent, with an evaluation ladder that holds the questions fixed while varying only the evidence supplied to the answerer, from text-only probes through …
@misc{garg2026answering,
title = {{Answering Without Arriving: An Evidence Audit for Embodied Question Answering}},
author = {Shubh Garg and Divyansh Pandey and Niyati Rawal and Vedant Khandelwal and Amit Sheth},
year = {2026},
eprint = {2611.01958},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.01958},
}