cs.RO（2025-03-06）

📊 共 35 篇论文 | 🔗 11 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (27 🔗8) 支柱二：RL算法与架构 (RL & Architecture) (4 🔗1) 支柱三：空间感知与语义 (Perception & Semantics) (3 🔗2) 支柱四：生成式动作 (Generative Motion) (1)

🔬 支柱一：机器人控制 (Robot Control) (27 篇)

#题目一句话要点标签🔗⭐
1 Refined Policy Distillation: From VLA Generalists to RL Experts 提出精炼策略蒸馏(RPD)，提升VLA模型在机器人操作任务中的性能。 manipulation reinforcement learning distillation
2 Whole-Body Model-Predictive Control of Legged Robots with MuJoCo 提出基于MuJoCo和iLQR的全身模型预测控制方法，实现四足和人形机器人的实时控制。 quadruped legged robot humanoid ✅
3 Towards Autonomous Reinforcement Learning for Real-World Robotic Manipulation with Large Language Models 提出ARCHIE，利用LLM自主生成奖励函数，实现真实机器人操作的自主强化学习。 manipulation bi-manual motion planning
4 VLA Model-Expert Collaboration for Bi-directional Manipulation Learning 提出VLA模型-专家协作框架，提升机器人操作任务泛化性与人机交互效率 manipulation vision-language-action VLA ✅
5 Learning Generalizable Language-Conditioned Cloth Manipulation from Long Demonstrations 提出基于LLM分解的长程cloth操作学习框架，提升泛化性 manipulation imitation learning large language model
6 PALo: Learning Posture-Aware Locomotion for Quadruped Robots PALo：一种用于四足机器人的姿态感知运动学习方法，提升复杂地形的敏捷性和鲁棒性。 quadruped locomotion sim-to-real
7 Modeling Dynamic Hand-Object Interactions with Applications to Human-Robot Handovers 提出基于物理模拟和强化学习的动态抓取合成方法，并应用于人机协作 manipulation bi-manual bimanual manipulation
8 GRaD-Nav: Efficiently Learning Visual Drone Navigation with Gaussian Radiance Fields and Differentiable Dynamics GRaD-Nav：利用高斯辐射场和可微动力学高效学习视觉无人机导航 sim-to-real reinforcement learning deep reinforcement learning ✅
9 QuietPaw: Learning Quadrupedal Locomotion with Versatile Noise Preference Alignment QuietPaw：通过多功能噪声偏好对齐学习四足运动 quadruped locomotion
10 Music-Driven Legged Robots: Synchronized Walking to Rhythmic Beats 提出音乐驱动的四足机器人控制框架，实现与节奏同步的行走 quadruped legged robot locomotion ✅
11 Curiosity-Driven Imagination: Discovering Plan Operators and Learning Associated Policies for Open-World Adaptation 提出基于好奇心驱动想象的混合规划学习系统，用于开放世界机器人操作任务 manipulation motion planning world model
12 Data-Efficient Learning from Human Interventions for Mobile Robots PVP4Real：一种基于在线人工干预的移动机器人高效学习方法 quadruped sim-to-real reinforcement learning ✅
13 Neural Configuration-Space Barriers for Manipulation Planning and Control 提出基于神经配置空间障碍的操纵规划与控制方法，提升动态环境下的安全性与效率。 manipulation motion planning
14 SeGMan: Sequential and Guided Manipulation Planner for Robust Planning in 2D Constrained Environments 提出SeGMan，解决2D约束环境下鲁棒的序列化操作规划问题 manipulation motion planning
15 Shaken, Not Stirred: A Novel Dataset for Visual Understanding of Glasses in Human-Robot Bartending Tasks 提出GlassNICOLDataset，用于提升人机交互调酒任务中机器人对玻璃杯的视觉理解能力。 humanoid humanoid robot open-vocabulary
16 SAFE-TAXI: A Hierarchical Multi-UAS Safe Auto-Taxiing Framework with Runtime Safety Assurance and Conflict Resolution 提出SAFE-TAXI框架，解决多无人机地面自动滑行中的安全保障与冲突消解问题 MPC motion planning spatiotemporal
17 DogLegs: Robust Proprioceptive State Estimation for Legged Robots Using Multiple Leg-Mounted IMUs DogLegs：一种利用多腿部IMU的鲁棒腿式机器人本体状态估计方法 legged robot ✅
18 Occlusion-Aware Consistent Model Predictive Control for Robot Navigation in Occluded Obstacle-Dense Environments 提出一种遮挡感知一致性模型预测控制，用于解决遮挡密集环境下的机器人导航问题。 model predictive control
19 Manipulation of Elasto-Flexible Cables with Single or Multiple UAVs 提出多无人机操控弹性电缆的新方法以解决操控难题 manipulation
20 The Spinning Blimp: Design and Control of a Novel Minimalist Aerial Vehicle Leveraging Rotational Dynamics and Locomotion 提出一种新型旋转飞艇，利用旋转动力学和运动实现低能耗稳定飞行 locomotion
21 Planning and Control for Deformable Linear Object Manipulation 提出一种高效易部署的DLO操作框架，结合全局规划和局部控制解决复杂环境下的DLO操作问题。 manipulation
22 Energy Consumption of Robotic Arm with the Local Reduction Method 提出局部缩减法，优化机器人手臂能耗，提升工业自动化可持续性。 MPC model predictive control
23 A Convex Formulation of Material Points and Rigid Bodies with GPU-Accelerated Async-Coupling for Interactive Simulation 提出基于GPU加速异步耦合的凸优化MPM刚体交互模拟方法 manipulation
24 GRIP: A General Robotic Incremental Potential Contact Simulation Dataset for Unified Deformable-Rigid Coupled Grasping GRIP：用于统一可变形-刚性耦合抓取的通用机器人增量势接触模拟数据集 manipulation
25 ExoNav II: Design of a Robotic Tool with Follow-the-Leader Motion Capability for Lateral and Ventral Spinal Cord Stimulation (SCS) 提出一种具有随动功能的机器人工具，用于脊髓侧方和腹侧刺激 locomotion
26 ViT-VS: On the Applicability of Pretrained Vision Transformer Features for Generalizable Visual Servoing ViT-VS：利用预训练Vision Transformer特征实现通用视觉伺服 manipulation ✅
27 SRSA: Skill Retrieval and Adaptation for Robotic Assembly Tasks SRSA：面向机器人装配任务的技能检索与自适应框架 manipulation ✅

🔬 支柱二：RL算法与架构 (RL & Architecture) (4 篇)

#题目一句话要点标签🔗⭐
28 Dexterous Grasping with Real-World Robotic Reinforcement Learning DexGraspRL：一种基于真实世界机器人强化学习的灵巧抓取方法 reinforcement learning imitation learning affordance
29 RA-DP: Rapid Adaptive Diffusion Policy for Training-Free High-frequency Robotics Replanning 提出RA-DP，实现机器人高频自适应扩散策略，无需训练即可应对动态环境下的重规划问题。 diffusion policy motion adaptation
30 Robotic Compliant Object Prying Using Diffusion Policy Guided by Vision and Force Observations 提出基于视觉与力觉引导的扩散策略，解决机器人柔顺物体撬取难题 diffusion policy zero-shot transfer ✅
31 OPG-Policy: Occluded Push-Grasp Policy Learning with Amodal Segmentation OPG-Policy：利用非模态分割学习的遮挡物体推抓策略 policy learning

🔬 支柱三：空间感知与语义 (Perception & Semantics) (3 篇)

#题目一句话要点标签🔗⭐
32 EvidMTL: Evidential Multi-Task Learning for Uncertainty-Aware Semantic Surface Mapping from Monocular RGB Images 提出EvidMTL以解决不确定性语义表面映射问题 depth estimation scene understanding semantic mapping
33 Real-time Spatial-temporal Traversability Assessment via Feature-based Sparse Gaussian Process 提出基于特征的稀疏高斯过程时空可通行性评估方法，用于复杂地形下的机器人自主导航。 traversability ✅
34 MarsLGPR: Mars Rover Localization with Ground Penetrating Radar 提出基于探地雷达的火星车定位方法MarsLGPR，提升高滑移环境下的定位精度。 visual odometry feature matching ✅

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
35 GAGrasp: Geometric Algebra Diffusion for Dexterous Grasping GAGrasp：利用几何代数扩散模型生成灵巧抓取 physically plausible

⬅️ 返回 cs.RO 首页 · 🏠 返回主页