cs.RO（2026-10-07）

📊 共 48 篇论文 | 🔗 9 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (35 🔗6) 支柱二：RL算法与架构 (RL & Architecture) (6 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (4 🔗2) 支柱三：空间感知与语义 (Perception & Semantics) (1) 支柱八：物理动画 (Physics-based Animation) (1) 支柱五：交互与反应 (Interaction & Reaction) (1)

🔬 支柱一：机器人控制 (Robot Control) (35 篇)

#题目一句话要点标签🔗⭐
1 HuMBLE: Human Motion-Driven Behavior Learning for Embodied Locomotion 提出HuMBLE框架以解决人形机器人运动控制的通用性问题 humanoid humanoid robot humanoid locomotion
2 Energy-Efficient Gait Adaptation via Hierarchical Reinforcement Learning for Quadrupedal Locomotion Across Diverse Terrains 提出分层强化学习框架以提高四足机器人在多种地形上的能效 quadruped legged robot locomotion
3 YUBI-STAG: Contact and Semantic-Rich Alignment for VLAs via Automated Video-Language Grounding 提出YUBI-STAG框架以解决VLA模型与细粒度语言对齐问题 manipulation bi-manual contact-aware
4 RoboPace: Contact-Aware Time-Optimal Retiming for Action-Chunk Policies 提出RoboPace以解决机器人动作时序优化问题 manipulation dual-arm teleoperation
5 Precise SE(3) End-Effector Tracking in Whole-Body Humanoid Control 提出ResGAC以解决人形机器人精确末端执行器跟踪问题 humanoid humanoid control locomotion ✅
6 Immiscible Diffusion Policy: Preserving Multimodal Robot Actions through Label-Free Noise Assignment 提出不可混溶扩散策略以解决机器人多模态动作保留问题 humanoid manipulation diffusion policy
7 TempoBridge: Language-Guided Tempo Control for Vision-Language-Action Policies 提出TempoBridge以解决视觉-语言-动作模型的节奏控制问题 manipulation vision-language-action VLA
8 Agentic RSR: Real-to-Sim-to-Real through Scene Reconstruction and Execution-Grounded Robot Policies 提出Agentic RSR框架以解决真实机器人任务执行中的场景重建与策略开发问题 manipulation sim-to-real MAE
9 Many Ways to Succeed: Diversity-Driven RL Fine-Tuning for VLA Generalization 提出DRIVE以解决VLA政策泛化不足问题 dual-arm reinforcement learning vision-language-action
10 RobotWorld: Benchmarking Multimodal Agents for Robot Use Across Diverse Tasks and Embodiments 提出RobotWorld以评估多模态智能体在机器人任务中的表现 locomotion manipulation mobile manipulation
11 Co${}^{2}$Skill: Whole-Body Control via Skill Composition for Long-Horizon Human-Environment Interaction 提出Co²Skill以解决复杂环境中的全身控制问题 whole-body control manipulation loco-manipulation
12 LLA-MPPI: Rapidly Adaptive Whole-body Control of Legged Robots with GPU-Accelerated Parallel Simulations 提出LLA-MPPI以解决腿部机器人适应性控制问题 legged robot whole-body control Unitree
13 Benchmarking Behavioral Steerability in Behavior Foundation Models 提出RoboSteer基准以评估行为基础模型的可操控性 humanoid foundation model multimodal
14 Contact-Aware Imitation Learning Through Contact Factorization 提出FACE框架以解决机器人接触条件变化下的模仿学习问题 manipulation imitation learning contact-aware
15 Sparse Feature Policy Unlearning Mitigates State Hallucination in Vision-Language-Action Models 提出SOUL以解决视觉-语言-动作模型中的状态幻觉问题 manipulation vision-language-action VLA
16 CMP-IRRT*: A Perception-Assisted Height-Adaptive Planner for Quadruped Robots 提出CMP-IRRT*以解决四足机器人低障碍物规划问题 quadruped Unitree Mamba ✅
17 RoboRender: Robot-Oriented Video Generation for Visual Sim-to-Real Transfer 提出RoboRender以解决视觉仿真与现实转移问题 manipulation mobile manipulation sim-to-real ✅
18 Factorized Tactile Representation and Control for Sim-to-Real Manipulation 提出因子化触觉表示与控制框架以解决仿真到现实的操作问题 manipulation sim-to-real
19 RoboQuest: Generalist Physical Agents that Search, Inspect and Test 提出RoboQuest以解决物理智能体在不确定环境中的探索问题 manipulation mobile manipulation foundation model
20 Distributed Motion Planning for Multi-Robot Systems under Topological Constraints 提出基于MPC的分布式控制方法以解决多机器人系统的拓扑约束问题 MPC model predictive control motion planning
21 RobotAPO: Adversarial Physics Preference Optimization for Robotic Manipulation Video Generation 提出RobotAPO以解决机器人操作视频生成中的物理一致性问题 manipulation flow matching penetration
22 Towards Accurate End-Effector Localization for UMI-Style Robotic Manipulation Teaching 提出MILD和AprilVINS以解决机器人操作教学中的末端执行器定位问题 manipulation policy learning VIO
23 Point It, Strike It: Direction-Conditioned Dynamic Manipulation of Deformable Linear Objects 提出TRACE与RECAP以解决可变形线性物体的动态操控问题 manipulation sim-to-real flow matching
24 Targeted Modality Dropout for Real-Robot Manipulation Robust to Intermittent Vision Loss 提出针对模态丢失的模态选择机制以增强机器人操作鲁棒性 manipulation bi-manual imitation learning
25 Predicted Futures Are Not Enough: Learning Executable Goals for Robot Manipulation 提出Entity-Level Goal Readout以解决机器人操控中的目标执行问题 manipulation world model world models ✅
26 Borrowed Eyes: Markerless Nano-UAV Flight with an Active Quadruped Observer 提出基于四足机器人观察的无标记纳米无人机定位方法 quadruped model predictive control
27 RFPO: Rectified Flow Policy Optimization for Embodied Control 提出RFPO以解决流政策在机器人控制中的执行效率问题 locomotion Unitree policy learning ✅
28 Temporal Visuo-Tactile Learning for Dexterous Grasp Stability 提出动态触觉感知以提升灵巧抓取稳定性 manipulation dexterous manipulation multimodal ✅
29 RoboPrompt: Intuitive Robot Policy Steering with Sparse Human Input 提出RoboPrompt以解决机器人政策引导的稀疏人类输入问题 teleoperation imitation learning diffusion policy
30 OpenViTac: Learning and Benchmarking Visuo-Tactile Policies in a Unified Sim-and-Real Framework 提出OpenViTac以解决机器人触觉反馈评估统一性问题 manipulation policy learning VLA
31 Long-WAM: Scaling the Context of World-Action Models 提出Long-WAM以解决实时机器人控制中的历史信息利用问题 manipulation Unitree egocentric
32 LACE-CRAFT: Robot Co-Design with Actor Inheritance and Blackboard Collaboration 提出LACE-CRAFT以解决机器人共设计中的控制经验丢失问题 locomotion policy learning
33 Enhancing Robotic Perception and Adaptability through Sensor Fusion and Origami-Inspired Designs 提出一种基于传感器融合和折纸设计的机器人感知与适应性增强方法 locomotion
34 ClimbLab: MATLAB Simulation Platform for Legged Climbing Robotics 提出ClimbLab以支持腿足攀爬机器人模拟 quadruped
35 Adaptive Risk-Certified Event-Triggered Replanning for Dynamic Navigation 提出CERT-Replan以解决动态环境下安全导航问题 MPC

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

#题目一句话要点标签🔗⭐
36 Juno: Taming Predictive Latents for Vision-Language-Action Models 提出Juno框架以解决VLA模型中的预测潜变量问题 policy learning world model world models
37 TMT: Runtime Backdoor Detection for Vision-Language-Action Policies on Unseen Tasks 提出TMT以解决视觉-语言-动作策略中的后门检测问题 latent dynamics distillation vision-language-action ✅
38 ΔWAM: Distilling Action Tangent Fields into World Action Models 提出ΔWAM以提高机器人在复杂环境中的动作预测能力 world model world models world action model
39 RealtimeWAM: How Fast Can I Run My World Action Model? 提出RealtimeWAM以解决世界动作模型推理延迟问题 world action model world action models
40 Decoding Neural Population Dynamics through Robotic Analog 提出机器人类比以解码神经群体动态与运动结果的关系 reinforcement learning Eureka multimodal
41 Adaptive Code Generation for Controlling Robots 提出自适应代码生成方法以解决机器人控制中的意图表达问题 distillation large language model

🔬 支柱九：具身大模型 (Embodied Foundation Models) (4 篇)

#题目一句话要点标签🔗⭐
42 Rephrase Before You Act: Characterizing and Mitigating Language Sensitivity in Vision-Language-Action Models 提出重述策略以缓解视觉语言行动模型的语言敏感性问题 vision-language-action VLA large language model ✅
43 Do Vision-Language-Action Models Understand Instructions? A Mechanistic Interpretability Study on Language Grounding 通过机制可解释性研究提升视觉-语言-动作模型的指令理解能力 vision-language-action VLA
44 MultiFly: A Real-World Multimodal Aerial Dataset with Annotation-Efficient Label Transfer and Cross-Modal Semantic Consistency 提出MultiFly数据集以解决多模态语义感知问题 multimodal ✅
45 End-to-End Autonomous Generation of Human Assembly Plans 提出端到端方法自动生成人类装配计划以解决手动生成问题 large language model multimodal

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

#题目一句话要点标签🔗⭐
46 NeRFifyMesh: Optimizing Neural Radiance Fields from Textured Meshes for Robotics Scene Building 提出NeRFifyMesh以解决机器人场景构建中的NeRF模型转换问题 NeRF neural radiance field semantic mapping

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

#题目一句话要点标签🔗⭐
47 AirGroundVLN: A Large-Scale Benchmark for Goal-Oriented Air-Ground Collaborative Vision-and-Language Navigation 提出AirGroundVLN以解决目标导向的空地协作视觉语言导航问题 spatiotemporal VLN

🔬 支柱五：交互与反应 (Interaction & Reaction) (1 篇)

#题目一句话要点标签🔗⭐
48 COOL: Curiosity-Driven Object Ownership Learning for Personalized Robotic Assistance 提出COOL框架以解决个性化机器人服务中的物品归属问题 human-object interaction

⬅️ 返回 cs.RO 首页 · 🏠 返回主页