cs.RO（2026-10-15）

📊 共 65 篇论文 | 🔗 8 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (47 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (11 🔗3) 支柱九：具身大模型 (Embodied Foundation Models) (3 🔗1) 支柱三：空间感知与语义 (Perception & Semantics) (1) 支柱七：动作重定向 (Motion Retargeting) (1) 支柱六：视频提取与匹配 (Video Extraction) (1) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱一：机器人控制 (Robot Control) (47 篇)

#题目一句话要点标签🔗⭐
1 TRACE: Typed Reference-Aware Conversational Editing for Humanoid Motion Diffusion 提出TRACE框架以解决人形机器人运动生成与编辑问题 humanoid humanoid robot Unitree
2 KODA: Kinematics--Observation Data Alignment for Humanoid Learning from Human Egocentric Demonstration 提出KODA以解决人类自我中心演示数据对齐问题 humanoid whole-body control manipulation
3 VeriVLA: Adaptive Test-Time Verification for Flow-Based VLAs with Latent World Models 提出VeriVLA以解决流基视觉-语言-动作策略的验证问题 manipulation world model world models
4 MAIA: Anticipating Interaction Anomalies in Contact-Rich Manipulation via Multimodal World Models 提出MAIA框架以解决接触丰富操作中的异常预测问题 manipulation world model world models
5 Inertially Significant Humanoid Teleoperation via Whole Body Model Predictive Control and Online Parameter Estimation 通过全身模型预测控制与在线参数估计解决人形机器人动态平衡问题 humanoid humanoid robot MPC
6 Fast-VLA: You Only Think Once for Robot Manipulation 提出Fast-VLA以提高机器人操控效率 manipulation vision-language-action VLA ✅
7 Cross-Embodiment Causal Transfer from Human Videos to Robot Manipulation 提出Causal-Beam-Guided Subtask Transfer以解决机器人操作数据不足问题 manipulation cross-embodiment human motion
8 Evo-DexBench: Benchmarking Generalization across Related Tasks for Dexterous Manipulation 提出Evo-DexBench以解决灵巧操控任务间知识迁移问题 manipulation dexterous manipulation teleoperation
9 Compliant GMT: Making General Motion Tracking Compliant via Residual Reference Adaptation 提出C-GMT框架以解决通用运动跟踪中的合规性问题 humanoid humanoid robot humanoid control
10 CQDC-Warp: GPU-Accelerated Differentiable Quasi-Dynamic Simulation for Contact-Rich Manipulation 提出CQDC-Warp以解决接触丰富的机器人操作轨迹优化问题 whole-body control manipulation bi-manual
11 Spark-0: Scaling Spatially Aligned Tactile Pretraining for Robot Dexterous Manipulation 提出Spark-0以解决机器人灵巧操作中的触觉感知问题 humanoid manipulation dexterous manipulation
12 FastVLA: Task-Conditioned and Temporally Adaptive Visual Token Compression for Efficient Vision-Language-Action Models 提出FastVLA以解决视觉-语言-动作模型的高效推理问题 manipulation vision-language-action VLA ✅
13 MuSTMM: Learning Sound-Triggered Mobile Manipulation across Multiple Scenes 提出MuSTMM以解决多场景下声音触发的移动操控问题 manipulation mobile manipulation egocentric
14 Dexterity from Supervised Play 提出DSP框架以解决机器人自主学习灵巧操作问题 manipulation dexterous manipulation sim-to-real
15 Whole-Body Dexterity 提出全身灵巧性框架以解决人形机器人操作局限问题 humanoid whole-body control manipulation
16 Trajectory-based Diffusion from One-shot Human Video for Generalizable Manipulation 提出基于轨迹的扩散方法以解决人类视频一-shot操控问题 manipulation teleoperation imitation learning
17 Sim-to-Real Reinforcement Learning for Path-Safe Target Reaching in Robotic Harvesting of Hydroponic Lettuce 提出PSA-PPO以解决水培生菜机器人采摘中的路径安全问题 manipulation sim-to-real reinforcement learning
18 Policy-Aligned Predictive Verification for Adaptive VLA Action-Chunk Execution 提出政策对齐的预测验证方法以优化VLA动作执行 manipulation vision-language-action VLA
19 DemoStream: Batched Task-and-Motion-Plan Optimization for Automated Learning from Demonstration 提出DemoStream以解决自动化学习示范中的演示质量问题 manipulation motion planning teleoperation
20 TS-WAM: Temporal Latent Actions and World-Action Coupling for Vision-Language Robot Control 提出TS-WAM以解决视觉-语言机器人控制中的未来状态预测问题 manipulation world model world models
21 SPEAR: Selective Preference Elicitation with Adaptive Rewards for VLM-Guided Robotic Manipulation 提出SPEAR框架以解决VLM引导的机器人操作中的偏好标注效率问题 manipulation Aloha Mobile Aloha
22 Training-Free Target-Aware Refinement of 3D Gaussian Language Map 提出无训练目标感知的3D高斯语言图优化方法以解决定位问题 legged robot 3D gaussian splatting gaussian splatting ✅
23 LTM: Latent Transition Based In-Context Memory for Vision-Language-Action Models 提出潜在转移记忆以解决机器人操控中的记忆问题 manipulation vision-language-action
24 LIBERO-PeRM: Benchmarking Personalized Robotic Manipulation 提出LIBERO-PeRM以解决个性化机器人操控基准问题 manipulation preference learning VLA ✅
25 Safe and Fuel-Efficient Motion Planning for Lunar Propulsive Drones 提出一种安全且高效的运动规划框架以解决月球推进无人机的路径规划问题 MPC model predictive control motion planning
26 MOSS: Mode-Oriented Source Shaping for Robot Flow Matching Policies 提出MOSS以解决机器人流匹配策略中的多模态选择问题 manipulation flow matching multimodal
27 Uni-PNP: Unified Prehensile and Non-prehensile Policy Learning for Generalizable Object Pose Reaching 提出Uni-PNP以解决物体姿态到达中的抓取与非抓取动作整合问题 manipulation reinforcement learning policy learning
28 Diffeomorphic Hidden-State Steering: Improving Imitation Learning Policy Robustness through Expert Hidden-State Geometry 提出微分隐状态引导以提升模仿学习策略的鲁棒性 manipulation imitation learning transformer policy
29 CarryOn: A Dual-Loop Harness for Personalized and Reliable Robot Assistance 提出CarryOn以解决个性化和可靠性不足的机器人辅助问题 manipulation vision-language-action VLA
30 Manifold Generative Annealing for Contact Constraint-aware Impedance Control with Reinforcement Learning Prior 提出流形生成退火方法以解决接触约束下的阻抗控制问题 humanoid reinforcement learning
31 Mem4Act: Learning Effective Latent Memory for Long-Horizon Robotic Manipulation 提出Mem4Act以解决长时间机器人操控中的记忆问题 manipulation
32 Auditable and Deployable Multi-Agent Motion Planning via Posetal Games from Legal Guidelines 提出基于法律指南的多智能体运动规划框架以解决合规性问题 motion planning
33 MR-MPCC: Model Predictive Contouring Control for Robotic Manipulator Tracking under Moving Reference Frames 提出MR-MPCC以解决机器人在移动参考框架下的路径跟踪问题 manipulation MPC model predictive control
34 Value as Memory: Online Value-Guided Post-training for General Robot Policies 提出Value as Memory框架以解决机器人政策在线后训练问题 manipulation reinforcement learning offline RL
35 Attribution-Gated Joint Estimation of Intention and Capability for Assist-as-Needed Upper-Limb Robots 提出基于归因门控的联合估计方法以解决上肢机器人辅助控制问题 MPC multimodal
36 AVALON: Zero-Shot Indoor Navigation with Human-Seeded Topological Memory 提出AVALON框架以解决室内导航中的人机协作问题 Unitree traversability
37 CoDA: Co-Evolution of Data Engines and Action Policies 提出CoDA框架以解决机器人训练环境构建问题 manipulation
38 Sub-Millisecond Reflexes for Continuous Physical Human-Robot Interaction 提出亚毫秒反应引擎以解决人机交互延迟问题 manipulation
39 Hierarchical Building-Floo-Room Framework for Privacy-Preserving Autonomous Delivery in Large-Scale Buildings 提出分层建筑-楼层-房间框架以解决大型建筑隐私保护自主配送问题 reachability-aware
40 CLIS-MPPI: Closed-Loop Model Predictive Path Integral Control With Initial-State Sampling Under Execution Delay 提出CLIS-MPPI以解决执行延迟导致的状态不匹配问题 motion planning
41 Vision-Guided Closed-Loop Autonomous Navigation of Light-Driven Janus Microbots 提出视觉引导闭环自主导航框架以解决微米级机器人导航挑战 manipulation
42 LIN-SLAM: Lightweight LiDAR--Inertial Neural SLAM for Real-Time Onboard Navigation 提出LIN-SLAM以解决实时导航中的SLAM精度与效率问题 Unitree
43 MAIA: Anticipating Interaction Anomalies in Contact-Rich Manipulation via Multimodal World Models 提出MAIA框架以解决接触丰富操作中的异常预测问题 manipulation world model world models
44 Embodiment-Classifier Guidance: Transferring Frozen VLAs Across Robot Dynamics without Demonstrations 提出嵌入分类器引导以解决机器人动态转移问题 sim-to-real VLA
45 FCTS: Failure-Guided Curriculum Tree Search Over Generated Terrain for Legged Parkour 提出FCTS以解决四足机器人在复杂地形中的训练问题 quadruped locomotion parkour
46 Scaling Online Exploration for Robot Agents 提出RAX框架以解决机器人在线探索能力不足的问题 humanoid manipulation mobile manipulation
47 Has GPT-6-Astra Achieved Embodied Intelligence? 提出GPT-6-Astra以挑战传统机器人智能模型构建 humanoid manipulation dexterous hand

🔬 支柱二：RL算法与架构 (RL & Architecture) (11 篇)

#题目一句话要点标签🔗⭐
48 What Should Robot Policies Learn to Predict? Rethinking Supervision Targets in Vision-Language-Action Learning 提出统一的VLA框架以优化机器人政策的监督目标 policy learning JEPA vision-language-action
49 Multi-Teacher On-Policy Distillation for Flow-Based Vision-Language-Action Models 提出多教师在线蒸馏方法以解决流式视觉-语言-动作模型的技能整合问题 distillation vision-language-action VLA
50 REKI: Registering Knowledge in VLAs 提出REKI模型以优化视觉-语言-动作任务的性能与效率 flow matching egocentric vision-language-action ✅
51 From Optical Flow to Safety: Visual Safety Certification via Scale-Free Relative Motion 提出视觉安全认证方法以解决单目视觉导航中的碰撞风险问题 world model world models optical flow
52 T2-z: Test-Time Optimization of Environment Latents for Robotic World Models 提出T2-Z以优化机器人世界模型中的环境潜变量 world model world models latent optimization
53 Plans Are Not Executions: Adapting Frozen World-Model Planners to Controller-Induced Execution Shift 提出ExecutionBridge以解决规划与执行之间的差距问题 world model world models
54 GRID-Planner: Graph-Structured Relational Reward Shaping for Diffusion Planning 提出GRID-Planner以解决复杂多智能体环境下的自主驾驶规划问题 distillation reward shaping ✅
55 Asking Smarter Questions: Targeted Active Preference Learning for Underspecified Rewards 提出针对性主动偏好学习以解决奖励学习中的信息不足问题 preference learning
56 X-Fly: Learning to Reach and Stop in Full-Chunk Aerial Vision-Language Navigation 提出X-Fly以解决全块空中视觉语言导航中的停止问题 behavior cloning VLN
57 DriveWorld3D Evidence Constrained 3D World Modeling and Generative Simulation for Real-World Autonomous Driving 提出DriveWorld3D以解决自动驾驶中的三维世界建模问题 world model world models 3D gaussian splatting
58 When Plans Change, Remember: Replanning-Guided Visual Memory for Vision-Language-Action Models 提出RePlanMem-VLA以解决视觉语言行动模型中的记忆冗余问题 flow matching vision-language-action VLA ✅

🔬 支柱九：具身大模型 (Embodied Foundation Models) (3 篇)

#题目一句话要点标签🔗⭐
59 When to Ask for Help: A Decision-Theoretic Account and Measurement Standard for Slow-Module Scheduling in Dual-System Vision-Language-Action Models 提出决策理论框架以优化双系统视觉语言行动模型中的慢模块调度 vision-language-action VLA
60 A Flow-Theoretic Multimodal Control Framework Fusing Mental and Physical Engagement for Upper-Limb Rehabilitation 提出流理论多模态控制框架以解决上肢康复中的参与度问题 multimodal
61 An Agentic Framework for Proactive Robots with Procedural Assistance Memory 提出基于程序辅助记忆的主动机器人框架以提升用户协助体验 VLA ✅

🔬 支柱三：空间感知与语义 (Perception & Semantics) (1 篇)

#题目一句话要点标签🔗⭐
62 CF-Splat: Closed-Form Triangle Splatting with Mesh Priors 提出CF-Splat以解决现有三角形点云渲染效率低下问题 3DGS splatting

🔬 支柱七：动作重定向 (Motion Retargeting) (1 篇)

#题目一句话要点标签🔗⭐
63 Modular Inverse Dynamics and Wrench Recovery for Closed-Chain Mechanisms Using Path Assembly 提出模块化逆动力学与力矩恢复方法以解决闭链机制问题 motion reconstruction

🔬 支柱六：视频提取与匹配 (Video Extraction) (1 篇)

#题目一句话要点标签🔗⭐
64 VP-EVT: Active Visibility Preservation for Embodied Visual Tracking 提出VP-EVT以解决动态场景下的可见性保持问题 egocentric

🔬 支柱八：物理动画 (Physics-based Animation) (1 篇)

#题目一句话要点标签🔗⭐
65 RTISSLAM-3D: Full 6-DoF SLAM With an In-Air 3D Imaging Sonar 提出RTISSLAM-3D以解决传统声纳SLAM的高度信息丢失问题 PULSE

⬅️ 返回 cs.RO 首页 · 🏠 返回主页