cs.RO（2026-08-18）

📊 共 20 篇论文 | 🔗 1 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (12 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (3) 支柱三：空间感知与语义 (Perception & Semantics) (2) 支柱二：RL算法与架构 (RL & Architecture) (2) 支柱四：生成式动作 (Generative Motion) (1)

🔬 支柱一：机器人控制 (Robot Control) (12 篇)

#题目一句话要点标签🔗⭐
1 EATR-Stereo: Embodiment-Aware Routing of Paired Stereo Evidence for Humanoid Vision-Language-Action Control 提出EATR-Stereo以解决长时间人形机器人视觉-语言-行动控制问题 humanoid vision-language-action VLA
2 Hydra-0: Action Flow for Generalist World Modeling and Control 提出Hydra-0以解决通用世界建模与控制问题 shared control world model world models
3 CompCPZ: Preserving Multi-Modal Intent in Language-Guided Robot Manipulation 提出CompCPZ以解决语言引导机器人操作中的多模态意图保留问题 quadruped manipulation Unitree
4 MANIGUARD: A Benchmark and Data Suite for Specification-Grounded Safety Evaluation and Improvement of Robotic Manipulation 提出MANIGUARD框架以评估和提升机器人操作的安全性 manipulation motion planning teleoperation
5 Robust Brachiation on a Life-Sized Dual-Arm Robot Using Waypoint-Guided Reinforcement Learning 提出基于路径引导的强化学习方法以实现双臂机器人稳健的摆动行走 locomotion dual-arm sim-to-real
6 Iterative Grasp Pose Refinement: A Deep Reinforcement Learning Approach for 2D Vision 提出基于深度强化学习的抓取姿态迭代优化方法以解决机器人抓取问题 manipulation sim-to-real reinforcement learning
7 PRISM: Precision and contact-rich Real-world Industrial Skill dataset with Multimodal sensing 提出PRISM数据集以解决工业操作中的多模态感知问题 manipulation multimodal ✅
8 ORPA: Online Residual Policy Adaptation for Robot Manipulation Control with Human Feedback 提出ORPA框架以实时纠正机器人操作中的执行误差 manipulation imitation learning Aloha
9 HODAgent: Towards On-Demand, Responsive Humanoids for Physical World Human Interaction 提出HODAgent以解决人形机器人在服务场景中的响应性问题 humanoid humanoid robot Unitree
10 Effector-Centric NMPC of Tiltable-Multirotors for Offset-Free Omnidirectional Aerial Manipulation 提出基于效应器的NMPC框架以实现无偏差的全向空中操作 manipulation model predictive control
11 UniReflex: Plug-and-Play Force Control for Pretrained Generative Policies via Fast-Slow Reflex 提出UniReflex以解决生成策略缺乏闭环力控制的问题 bi-manual imitation learning
12 Optimal control of a swimming robot based on Purcell's microswimmer model 基于Purcell微游泳者模型的游泳机器人最优控制方法 locomotion

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

#题目一句话要点标签🔗⭐
13 Reuse Before You Retrieve: Diagnosing Headroom and Complementarity for Test-Time Augmentation of Embodied Multimodal Policies 提出测试时增强策略以优化多模态政策的行为选择 vision-language-action VLA multimodal
14 LIBERO-VIFO: Benchmarking the Capability and Safety of Visual Cue Following in Vision-Language-Action Models 提出LIBERO-VIFO以评估视觉线索跟随的能力与安全性 vision-language-action VLA
15 Embodied-Navigator: Point, Think, Memorize, and Align for Efficient Navigation 提出TAMP-Nav以解决现有视觉语言模型导航效率低下问题 chain-of-thought TAMP

🔬 支柱三：空间感知与语义 (Perception & Semantics) (2 篇)

#题目一句话要点标签🔗⭐
16 OVIP-SG: Open-Vocabulary Instance-Preserving Scene Graphs for Mapping and Retrieval of Small, Fine-Grained Objects 提出OVIP-SG以解决开放词汇感知下小物体映射与检索问题 semantic mapping semantic map open-vocabulary
17 Scalix: Uncertainty-Aware Scale-Consistent Monocular SLAM 提出Scalix以解决单目SLAM中的尺度模糊问题 monocular depth foundation model

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

#题目一句话要点标签🔗⭐
18 Prism-GRPO: Faster VLA Policy Optimization via Splitting Same-outcome Groups 提出Prism-GRPO以解决VLA策略优化中的采样效率问题 reinforcement learning PPO vision-language-action
19 Calibrated Predictive Safety for Heterogeneous Robots: An Action-Conditioned JEPA Framework with Model-Based Safety Shields 提出基于JEPA框架的异构机器人安全预测方法 world model world models JEPA

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
20 Physics-Informed Sliding-Window Particle Filtering for Tactile-Only In-Hand 6-DoF Object Pose Refinement 提出物理信息滑动窗口粒子滤波以解决触觉仅依赖的物体姿态精细化问题 penetration

⬅️ 返回 cs.RO 首页 · 🏠 返回主页