cs.RO（2025-09-16）

📊 共 35 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (18 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (9) 支柱三：空间感知与语义 (Perception & Semantics) (4 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (3) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱一：机器人控制 (Robot Control) (18 篇)

#题目一句话要点标签🔗⭐
1 Embracing Bulky Objects with Humanoid Robots: Whole-Body Manipulation with Reinforcement Learning 提出基于强化学习的人形机器人全身操作框架，解决拥抱大体积物体的稳定操作问题 humanoid humanoid robot manipulation
2 Contrastive Representation Learning for Robust Sim-to-Real Transfer of Adaptive Humanoid Locomotion 提出对比学习框架，提升人形机器人适应性步态的Sim-to-Real迁移鲁棒性 humanoid humanoid locomotion locomotion ✅
3 Integrating Trajectory Optimization and Reinforcement Learning for Quadrupedal Jumping with Terrain-Adaptive Landing 提出结合轨迹优化与强化学习的四足机器人跳跃框架，实现地形自适应着陆 quadruped locomotion trajectory optimization
4 \textsc{Gen2Real}: Towards Demo-Free Dexterous Manipulation by Harnessing Generated Video Gen2Real：利用生成视频实现免示教的灵巧操作 manipulation dexterous manipulation trajectory optimization
5 Collaborative Loco-Manipulation for Pick-and-Place Tasks with Dynamic Reward Curriculum 提出基于动态奖励课程的分层强化学习方法，解决单/双臂腿式机器人协同抓取放置任务。 legged robot manipulation loco-manipulation
6 StageACT: Stage-Conditioned Imitation for Robust Humanoid Door Opening StageACT：基于阶段条件模仿学习的鲁棒人形机器人开门方法 humanoid humanoid robot manipulation
7 Safety filtering of robotic manipulation under environment uncertainty: a computational approach 提出基于物理仿真的机器人操作安全过滤方案，应对环境不确定性 manipulation bi-manual bimanual manipulation
8 Hydrosoft: Non-Holonomic Hydroelastic Models for Compliant Tactile Manipulation 提出Hydrosoft模型，解决柔顺触觉操作中非完整约束下的水弹性建模问题 manipulation trajectory optimization
9 Model Predictive Control with Reference Learning for Soft Robotic Intracranial Pressure Waveform Modulation 提出基于参考学习的模型预测控制，用于软体机器人颅内压波形调节 MPC model predictive control
10 Empowering Multi-Robot Cooperation via Sequential World Models SeqWM：基于序列世界模型赋能多机器人协作 quadruped reinforcement learning world model
11 PerchMobi^3: A Multi-Modal Robot with Power-Reuse Quad-Fan Mechanism for Air-Ground-Wall Locomotion PerchMobi^3：一种用于空-地-墙运动的动力复用四风扇多模态机器人 locomotion
12 Object Pose Estimation through Dexterous Touch 提出基于灵巧触觉的主动探索物体姿态估计方法，解决视觉受限场景下的姿态估计问题。 manipulation bi-manual reinforcement learning ✅
13 Beyond Anthropomorphism: Enhancing Grasping and Eliminating a Degree of Freedom by Fusing the Abduction of Digits Four and Five SABD手：融合四指和五指外展/内收自由度，提升抓取能力并简化机械结构 manipulation dexterous manipulation reinforcement learning
14 Bridging Perception and Planning: Towards End-to-End Planning for Signal Temporal Logic Tasks 提出S-MSP，用于解决机器人信号时序逻辑任务的端到端感知与规划问题 motion planning task and motion planning
15 A Convex Formulation of Compliant Contact between Filaments and Rigid Bodies 提出一种基于凸优化的柔性细丝与刚体接触模拟框架，适用于软体机器人和形变物体操作。 manipulation PULSE
16 MoiréTac: A Dual-Mode Visuotactile Sensor for Multidimensional Perception Using Moiré Pattern Amplification MoiréTac：利用莫尔条纹放大实现多维感知的双模态视觉触觉传感器 manipulation dexterous manipulation
17 Leg-Arm Coordinated Operation for Curtain Wall Installation 针对幕墙安装，提出一种基于六足机器人腿-臂协同的分层优化控制框架 whole-body control
18 Underactuated Robotic Hand with Grasp State Estimation Using Tendon-Based Proprioception 提出基于肌腱本体感受的欠驱动机械手抓取状态估计方法 manipulation

🔬 支柱二：RL算法与架构 (RL & Architecture) (9 篇)

#题目一句话要点标签🔗⭐
19 TreeIRL: Safe Urban Driving with Tree Search and Inverse Reinforcement Learning TreeIRL：结合蒙特卡洛树搜索与逆强化学习的安全城市自动驾驶规划器 reinforcement learning imitation learning inverse reinforcement learning
20 GRATE: a Graph transformer-based deep Reinforcement learning Approach for Time-efficient autonomous robot Exploration 提出基于图Transformer的深度强化学习方法GRATE，提升机器人自主探索的时间效率。 reinforcement learning deep reinforcement learning DRL
21 ActiveVLN: Towards Active Exploration via Multi-Turn RL in Vision-and-Language Navigation ActiveVLN：基于多轮强化学习的主动探索视觉语言导航 reinforcement learning imitation learning reward shaping
22 An Uncertainty-Weighted Decision Transformer for Navigation in Dense, Complex Driving Scenarios 提出不确定性加权决策Transformer，提升复杂交通场景自动驾驶决策安全性。 decision transformer occupancy grid
23 Pre-trained Visual Representations Generalize Where it Matters in Model-Based Reinforcement Learning 预训练视觉表征显著提升模型强化学习在视觉域偏移下的泛化能力 reinforcement learning policy learning
24 Dense-Jump Flow Matching with Non-Uniform Time Scheduling for Robotic Policies: Mitigating Multi-Step Inference Degradation 提出Dense-Jump Flow Matching方法，通过非均匀时间调度优化机器人策略，缓解多步推理退化问题。 flow matching
25 VEGA: Electric Vehicle Navigation Agent via Physics-Informed Neural Operator and Proximal Policy Optimization VEGA：基于物理信息神经算子和PPO的电动汽车导航代理 reinforcement learning PPO teacher-student
26 ASTREA: Introducing Agentic Intelligence for Orbital Thermal Autonomy ASTREA：面向轨道热自主性的Agentic智能系统 reinforcement learning large language model
27 Robust Online Residual Refinement via Koopman-Guided Dynamics Modeling 提出KORR：利用Koopman引导的动态模型实现鲁棒的在线残差策略优化 policy learning imitation learning

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

#题目一句话要点标签🔗⭐
28 ROOM: A Physics-Based Continuum Robot Simulator for Photorealistic Medical Datasets Generation ROOM：用于生成逼真医学数据集的基于物理的连续体机器人模拟器 depth estimation monocular depth metric depth ✅
29 NavMoE: Hybrid Model- and Learning-based Traversability Estimation for Local Navigation via Mixture of Experts 提出NavMoE，通过混合专家模型实现机器人局部导航中更高效、泛化的地形可通行性估计。 traversability
30 Practical Handling of Dynamic Environments in Decentralised Multi-Robot Patrol 提出一种去中心化多机器人巡逻方法，用于动态环境下的持续监控。 traversability
31 UDON: Uncertainty-weighted Distributed Optimization for Multi-Robot Neural Implicit Mapping under Extreme Communication Constraints UDON：面向极端通信约束的多机器人神经隐式地图不确定性加权分布式优化 implicit representation

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

#题目一句话要点标签🔗⭐
32 The Better You Learn, The Smarter You Prune: Towards Efficient Vision-language-action Models via Differentiable Token Pruning LightVLA：通过可微Token剪枝提升视觉-语言-动作模型的效率与性能 vision-language-action VLA
33 Multi-Robot Task Planning for Multi-Object Retrieval Tasks with Distributed On-Site Knowledge via Large Language Models 提出基于LLM的多机器人任务规划框架，解决分布式知识下多目标检索任务 large language model
34 Toward Ownership Understanding of Objects: Active Question Generation with Large Language Model and Probabilistic Generative Model ActOwL：结合LLM与概率生成模型的主动问答，提升机器人对象所有权理解 large language model

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

#题目一句话要点标签🔗⭐
35 Spatiotemporal Calibration for Laser Vision Sensor in Hand-eye System Based on Straight-line Constraint 提出基于直线约束的时空标定方法，解决激光视觉手眼系统中时延和外参漂移问题 spatiotemporal

⬅️ 返回 cs.RO 首页 · 🏠 返回主页