cs.RO（2026-03-30）

📊 共 19 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (13 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (2) 支柱九：具身大模型 (Embodied Foundation Models) (2 🔗1) 支柱三：空间感知与语义 (Perception & Semantics) (1) 支柱六：视频提取与匹配 (Video Extraction) (1)

🔬 支柱一：机器人控制 (Robot Control) (13 篇)

#题目一句话要点标签🔗⭐
1 Cost-Matching Model Predictive Control for Efficient Reinforcement Learning in Humanoid Locomotion 提出基于代价匹配模型预测控制的人形机器人高效强化学习方法 humanoid humanoid locomotion locomotion
2 Active Stereo-Camera Outperforms Multi-Sensor Setup in ACT Imitation Learning for Humanoid Manipulation 主动双目相机在人形机器人操作的ACT模仿学习中优于多传感器配置 humanoid humanoid robot manipulation
3 ManipArena: Comprehensive Real-world Evaluation of Reasoning-Oriented Generalist Robot Manipulation ManipArena：面向推理的通用机器人操作真实世界综合评估框架 manipulation mobile manipulation world model
4 FocusVLA: Focused Visual Utilization for Vision-Language-Action Models FocusVLA：聚焦视觉利用的视觉-语言-动作模型，提升机器人操作性能 manipulation dexterous manipulation vision-language-action
5 Pandora: Articulated 3D Scene Graphs from Egocentric Vision Pandora：利用第一人称视觉构建可动3D场景图，提升机器人操作能力 manipulation mobile manipulation egocentric
6 DRIVE-Nav: Directional Reasoning, Inspection, and Verification for Efficient Open-Vocabulary Navigation 提出DRIVE-Nav以解决开放词汇目标导航中的路径效率问题 humanoid humanoid robot open-vocabulary ✅
7 Feel Robot Feels: Tactile Feedback Array Glove for Dexterous Manipulation 提出TAG触觉反馈手套系统，提升灵巧操作遥操作的性能与数据质量 manipulation dexterous manipulation teleoperation
8 Tac2Real: Reliable and GPU Visuotactile Simulation for Online Reinforcement Learning and Zero-Shot Real-World Deployment Tac2Real：用于在线强化学习和零样本迁移的可靠GPU视觉触觉模拟 manipulation sim-to-real reinforcement learning ✅
9 Tele-Catch: Adaptive Teleoperation for Dexterous Dynamic 3D Object Catching Tele-Catch：面向灵巧动态3D物体抓取的自适应遥操作框架 manipulation dexterous hand teleoperation
10 SOLE-R1: Video-Language Reasoning as the Sole Reward for On-Robot Reinforcement Learning SOLE-R1：利用视频-语言推理作为机器人强化学习的唯一奖励信号 manipulation reinforcement learning spatiotemporal
11 A Foldable and Agile Soft Electromagnetic Robot for Multimodal Navigation in Confined and Unstructured Environments 提出一种可折叠的软电磁机器人，用于狭窄非结构化环境下的多模态导航 locomotion multimodal
12 Dynamic Lookahead Distance via Reinforcement Learning-Based Pure Pursuit for Autonomous Racing 提出基于强化学习的Pure Pursuit算法，动态调整前瞻距离以优化自动驾驶赛车性能 sim-to-real reinforcement learning PPO
13 Serialized Red-Green-Gray: Quicker Heuristic Validation of Edges in Dynamic Roadmap Graphs 提出基于GPU加速的SerRGG框架，加速动态路标图中的边验证，提升机器人运动规划效率。 motion planning

🔬 支柱二：RL算法与架构 (RL & Architecture) (2 篇)

#题目一句话要点标签🔗⭐
14 StreamingVLA: Streaming Vision-Language-Action Model with Action Flow Matching and Adaptive Early Observation 提出StreamingVLA，通过异步并行化VLA各阶段，加速资源受限平台的视觉-语言-动作模型推理。 flow matching vision-language-action VLA
15 A Deep Reinforcement Learning Framework for Closed-loop Guidance of Fish Schools via Virtual Agents 提出基于深度强化学习的鱼群闭环引导框架，通过虚拟智能体实现鱼群控制。 reinforcement learning deep reinforcement learning PPO

🔬 支柱九：具身大模型 (Embodied Foundation Models) (2 篇)

#题目一句话要点标签🔗⭐
16 Fine-Tuning Large Language Models for Cooperative Tactical Deconfliction of Small Unmanned Aerial Systems 微调大语言模型，解决小型无人机系统协同战术冲突消解问题 large language model
17 Reducing Mental Workload through On-Demand Human Assistance for Physical Action Failures in LLM-based Multi-Robot Coordination REPAIR：基于LLM的多机器人协作中，通过按需人工辅助解决物理执行失败问题 large language model ✅

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

#题目一句话要点标签🔗⭐
18 osmAG-Nav: A Hierarchical Semantic Topometric Navigation Stack for Robust Lifelong Indoor Autonomy 提出osmAG-Nav，一种基于层级语义拓扑地图的室内自主导航系统，提升大规模环境下的导航鲁棒性。 occupancy grid traversability

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

#题目一句话要点标签🔗⭐
19 Point of View: How Perspective Affects Perceived Robot Sociability 研究视角对机器人社交性的影响：第一人称视角下社交性感知显著降低 egocentric

⬅️ 返回 cs.RO 首页 · 🏠 返回主页