cs.RO（2026-10-29）

📊 共 54 篇论文 | 🔗 8 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (35 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (6) 支柱九：具身大模型 (Embodied Foundation Models) (6 🔗2) 支柱三：空间感知与语义 (Perception & Semantics) (4) 支柱七：动作重定向 (Motion Retargeting) (2 🔗2) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱一：机器人控制 (Robot Control) (35 篇)

#题目一句话要点标签🔗⭐
1 AcrossDEX 1.0: From Human Motion Pretraining to Bimanual Dexterous Manipulation across Hand Embodiments 提出AcrossDEX 1.0以解决双手灵巧操作中的人机协调问题 manipulation dexterous manipulation bi-manual
2 TouchBridge: Tactile-Conditioned Human-to-Robot Transfer for Dexterous Manipulation 提出TouchBridge框架以解决人机触觉交互转移问题 manipulation dexterous manipulation human-to-robot
3 PRIME W0: Physical Priors from World Models for Embodied Manipulation 提出PRIME W0以解决VLA模型在长时序上下文处理中的不足 manipulation flow matching world model
4 LLM-guided Planar Loco-Manipulation for Quadruped Robots 提出基于LLM的平面运动操控方法以解决四足机器人在复杂环境中的导航问题 quadruped manipulation loco-manipulation ✅
5 ScheduleGen: Demonstration Generation via Skill Scheduling for Multi-Arm Imitation Learning 提出ScheduleGen以解决多臂机器人演示生成问题 humanoid humanoid robot manipulation
6 From Reasoning to Action: Consequence-Grounded Alignment for Vision-Language-Action Models 提出CoRAct框架以解决视觉-语言-行动模型的对齐问题 manipulation vision-language-action VLA ✅
7 Humanoid Dodgeball: Learning Fast Dynamic Obstacle Avoidance Motions for Humanoids 提出Humanoid Dodgeball框架以解决动态障碍物规避问题 humanoid reinforcement learning reward shaping
8 UMI-Live for In-The-Wild Bimanual Robotics Data Collection 提出UMI-Live以解决野外双手机器人数据收集问题 manipulation bi-manual teleoperation
9 Teaching a Tree to Recover: Online Distillation of an LLM Supervisor with Counterfactual Replay 提出在线蒸馏决策树以解决移动机器人规划问题 MPC model predictive control distillation
10 Learning Safe Flow Matching Robot Policies from Binary Feasibility Oracles 提出基于二元可行性预言机的安全流匹配机器人策略 manipulation policy learning flow matching
11 LOGIV: Graph-Verified Execution with Causal Local Repair for Long-Horizon Robot Tasks 提出LOGIV框架以解决长时间机器人任务的执行问题 dual-arm world action model world action models
12 SONIC-K1: Porting a Universal Humanoid Motion Tracker to a Small Humanoid on a Single GPU 提出SONIC-K1以实现小型人形机器人全身运动跟踪 humanoid motion tracking
13 EgoContact: Evaluating Physical Intelligence with Egocentric Video 提出EgoContact以评估物理智能中的接触点预测问题 manipulation affordance egocentric
14 SwapBot: Tool-Centric Robot Manipulation via Interchangeable End-Effectors and Tactile Sensing 提出SwapBot以解决工具中心机器人操作中的工具互换问题 manipulation teleoperation imitation learning ✅
15 RAIBO-Wagon: Real-Time Navigation System for a Legged Robot Towing a Wagon 提出RAIBO-Wagon以解决腿式机器人拖拽车厢的导航问题 quadruped legged robot flow matching
16 Tactus: Tactile & Pose Sensing Glove for Natural Dexterity Data Collection 提出Tactus手套以解决自然操作数据收集问题 manipulation dexterous manipulation world model
17 Improving Fine-Grained Manipulation through Demonstration Expansion and Execution Design 提出DEED框架以提升细粒度操作的执行效果 manipulation vision-language-action
18 Utility-WAM: Correcting Future-Induced Action Changes 提出Utility-WAM以解决未来引发的动作变化问题 manipulation bi-manual bimanual manipulation
19 Kinematic-Tactile Aligned Dexterous Retargeting with Contact Wrench Space Alignment 提出一种层次化框架以解决人机触觉信息转移问题 manipulation dexterous manipulation cross-embodiment
20 Event2Reward: Learning Robotic Rewards from Progress Relations 提出Event2Reward以解决机器人奖励学习中的进展关系问题 manipulation reinforcement learning policy learning
21 Gaze-Assisted Supervisory Control for Lunar Robotic Arm Teleoperation Under Concurrent Monitoring 提出注视辅助的监督控制框架以提升月球机器人臂遥控效率 teleoperation
22 A Tutorial on Shape Servoing of Deformable Objects-Part I: Fundamentals 提出基于数据驱动的形状控制方法以操控可变形物体 manipulation
23 DAWN: Diffusion-based Alignment and unfastening Workflow Network 提出DAWN以解决机器人自主拆卸任务的挑战 manipulation
24 MUSBRIC: LEGO-Compatible Pneumatic Muscle Bricks for Reconfigurable Musculoskeletal Robots 提出MUSBRIC以实现可重构的肌肉骨骼机器人 quadruped
25 Language-Grounded Force/Torque Pretraining for Extrinsic Contact Perception 提出语言引导的力/扭矩预训练以解决外部接触感知问题 manipulation
26 CuBot-STT: Simultaneously Transport and Tighten for Lattice-type Modular Reconfiguration Robot 提出CuBot-STT以解决模块化机器人运输与紧固问题 manipulation
27 Tracking Task Progress Through Errors and Recovery:A Benchmark and Recurrent Distributional Model 提出ReSTEP数据集与REMORA模型以解决机器人任务进度评估问题 manipulation
28 From Teleoperation to Long-Horizon Dexterity: Evaluating Visuomotor Skills and Offline Proxies for Dexterous Manipulation 提出新方法以提升机器人模仿学习的真实世界表现 manipulation dexterous manipulation teleoperation
29 CaPtain: A Pipeline for Post-Training Code-as-Policy Robot Agents 提出CaPtain以解决小型开放视觉语言模型的机器人操作问题 manipulation reinforcement learning vision-language-action
30 WorldBridge: Bridging Video and Action for Task Generalization in World Action Models 提出WorldBridge以解决机器人学习中的任务泛化问题 manipulation flow matching world action model
31 STAR: Steering Action Representations during Inference 提出STAR方法以解决视觉语言动作模型的晚层基础差距问题 manipulation vision-language-action VLA ✅
32 LOGIV: Graph-Verified Execution with Causal Local Repair for Long-Horizon Robot Tasks 提出LOGIV框架以解决长时间机器人任务中的执行可靠性问题 dual-arm world action model world action models
33 Contact Imitation Interface: Transferring Dexterous Skills by Showing How Contact Unfolds 提出接触模仿接口以解决灵巧操作技能转移问题 manipulation dexterous manipulation
34 CuPIQP: A Batched and Differentiable QP Solver on GPUs for Robot Learning and Control 提出CuPIQP以解决大规模机器人学习中的QP求解问题 whole-body control model predictive control
35 MUSBRIC: LEGO-Compatible Pneumatic Muscle Bricks for Reconfigurable Musculoskeletal Robots 提出MUSBRIC以实现可重构的肌肉骨骼机器人 quadruped

🔬 支柱二：RL算法与架构 (RL & Architecture) (6 篇)

#题目一句话要点标签🔗⭐
36 Continuous-Action World Models: Learning Cross-Embodiment Spatiotemporal Dynamics Priors 提出连续动作世界模型以解决跨体现机器人动作生成问题 world model world models cross-embodiment
37 How Much Should a VLA Trust Its Reasoning? Step-Level Guidance as a Measured Action-Value Function 提出基于步级指导的VLA信任机制以优化决策过程 SAC classifier-free guidance vision-language-action
38 Assuring Learning-Enabled Cyber-Physical Systems with Foundation Models: A Neuro-Symbolic Agenda for Testing, Verification, and Runtime Assurance 提出基础模型保障栈以解决学习驱动的网络物理系统的验证问题 world model world models foundation model
39 JEDI: Joint-Embedding Demonstration Compression for In-Context Imitation Learning 提出JEDI以解决长多模态轨迹压缩问题 imitation learning JEPA multimodal
40 PASA-KV: Attacking Action-Facing Visual Routes in World Action Models 提出PASA-KV以解决世界行动模型中的视觉路径攻击问题 world action model world action models
41 Gaussian Distillation of Flow Policies 提出高斯蒸馏方法以提升流政策的推理效率 distillation multimodal

🔬 支柱九：具身大模型 (Embodied Foundation Models) (6 篇)

#题目一句话要点标签🔗⭐
42 BURGER: Benchmarking and Understanding Robotic Generalization via Embodied Drawing 提出BURGER基准以解决机器人多样化行为学习问题 vision-language-action VLA multimodal ✅
43 RoboSeed: Multidimensional Evaluation of Vision-Language-Action and World-Action Models: Capability, Resource Affordability, and Research Evidence 提出RoboSeed框架以多维度评估机器人行动模型 vision-language-action VLA
44 Gym-Anything for Robotics: Scaling physical simulations with digital agents 提出Gym-Anything框架以解决机器人仿真范围狭窄问题 embodied AI vision-language-action VLA ✅
45 Surrogate Uncertainty for Black-box Visual Grounding 提出外部代理不确定性模型以解决视觉定位不确定性问题 visual grounding
46 Ask to Evolve: A Unified Harness for Lifelong Embodied Navigation with Hybrid Action Space and Persistent Memory 提出EvoLife以解决自主导航与记忆整合问题 VLN multimodal
47 The Athlete as a Scientist of One: Agentic Bayesian Experimental Design for Individualized Physiology from Consumer Wearables 提出EpiCoach以实现个性化运动生理学的可审计科学 large language model

🔬 支柱三：空间感知与语义 (Perception & Semantics) (4 篇)

#题目一句话要点标签🔗⭐
48 Decoding Geometry, Not Tokens: Efficient VLMs for Robot Affordance and Trajectory Prediction 提出几何解码方法以解决机器人抓取与轨迹预测问题 affordance
49 SymAnt: Grounding Long-Horizon Action Anticipation with Adaptable Symbolic Priors 提出SymAnt以解决长时间序列动作预测问题 affordance spatial relationship
50 Decision-Level Error Attribution and Depth-Aware Safety Evaluation of Monocular RGB Human Tracking for Autonomous Robot Navigation 提出决策级错误归因方法以提升单目RGB人类跟踪安全性 Depth Anything
51 Don't Risk It!: Trajectory-Conditioned Distributional Risk for LiDAR-Only Crowd Navigation 提出DRIP框架以解决LiDAR环境下的拥挤导航问题 occupancy grid

🔬 支柱七：动作重定向 (Motion Retargeting) (2 篇)

#题目一句话要点标签🔗⭐
52 OpenFlyV2: An onboard vision-language-action model for aerial perception and navigation 提出OpenFlyV2以解决无人机自主导航与感知问题 spatial relationship vision-language-action multimodal ✅
53 OpenFlyV2: An onboard vision-language-action model for aerial perception and navigation 提出OpenFlyV2以解决无人机感知与导航的多模态融合问题 spatial relationship vision-language-action multimodal ✅

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

#题目一句话要点标签🔗⭐
54 PhysVLMM: Interaction-Conditioned Physical Belief Maps for Articulated Objects in Robot Scene Maps 提出PhysVLMM以解决机器人场景中物体交互问题 differentiable simulation

⬅️ 返回 cs.RO 首页 · 🏠 返回主页