cs.RO（2025-10-09）

📊 共 26 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (21 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (3 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (2)

🔬 支柱一：机器人控制 (Robot Control) (21 篇)

#题目一句话要点标签🔗⭐
1 DexMan: Learning Bimanual Dexterous Manipulation from Human and Generated Videos DexMan：从人类和生成视频中学习双手动灵巧操作 humanoid humanoid robot manipulation
2 Towards Proprioception-Aware Embodied Planning for Dual-Arm Humanoid Robots 提出Proprio-MLLM，增强双臂人形机器人具身规划的本体感知能力 humanoid humanoid robot dual-arm
3 Humanoid Everyday: A Comprehensive Robotic Dataset for Open-World Humanoid Manipulation Humanoid Everyday：一个面向开放世界人型机器人操作的综合数据集 humanoid humanoid robot locomotion
4 Probabilistically-Safe Bipedal Navigation over Uncertain Terrain via Conformal Prediction and Contraction Analysis 提出基于一致性预测和收缩分析的概率安全双足机器人导航方法，解决地形不确定性下的稳健行走问题。 bipedal biped locomotion
5 USIM and U0: A Vision-Language-Action Dataset and Model for General Underwater Robots 提出USIM和U0以解决水下机器人多任务智能问题 manipulation mobile manipulation vision-language-action
6 Don't Run with Scissors: Pruning Breaks VLA Models but They Can Be Recovered 提出GLUESTICK，用于恢复剪枝后VLA模型性能，提升机器人操作和导航安全性 manipulation vision-language-action VLA ✅
7 Whole Body Model Predictive Control for Spin-Aware Quadrupedal Table Tennis 提出基于全身模型预测控制的四足机器人乒乓球系统，实现高速高精度回球 quadruped legged robot MPC
8 Trajectory Conditioned Cross-embodiment Skill Transfer TrajSkill：基于轨迹条件的跨具身操作技能迁移框架 manipulation optical flow human-to-robot
9 NovaFlow: Zero-Shot Manipulation via Actionable Flow from Generated Videos NovaFlow：通过生成视频中的可执行光流实现机器人零样本操作 quadruped manipulation trajectory optimization
10 R2RGEN: Real-to-Real 3D Data Generation for Spatially Generalized Manipulation 提出R2RGen，用于生成真实3D数据，提升机器人空间泛化操作能力。 manipulation mobile manipulation sim-to-real
11 Executable Analytic Concepts as the Missing Link Between VLM Insight and Precise Manipulation 提出GRACE框架以解决机器人精确操作与语义理解之间的鸿沟 manipulation affordance embodied AI
12 IntentionVLA: Generalizable and Efficient Embodied Intention Reasoning for Human-Robot Interaction IntentionVLA：面向人机交互的可泛化高效具身意图推理框架 manipulation vision-language-action VLA
13 Adaptive Motion Planning via Contact-Based Intent Inference for Human-Robot Collaboration 提出基于接触意图推断的自适应运动规划，用于人机协作。 motion planning large language model
14 FastUMI-100K: Advancing Data-driven Robotic Manipulation with a Large-scale UMI-style Dataset FastUMI-100K：大规模UMI风格数据集，推进数据驱动的机器人操作学习 manipulation multimodal ✅
15 DexNDM: Closing the Reality Gap for Dexterous In-Hand Rotation via Joint-Wise Neural Dynamics Model 提出DexNDM，通过关节级神经动力学模型弥合灵巧手内旋转的现实差距 manipulation dexterous manipulation sim-to-real ✅
16 Differentiable Particle Optimization for Fast Sequential Manipulation 提出SPaSM以解决实时顺序机器人操作问题 manipulation trajectory optimization
17 DM1: MeanFlow with Dispersive Regularization for 1-Step Robotic Manipulation DM1：通过分散正则化的MeanFlow实现单步机器人操作，解决表示崩溃问题。 manipulation flow matching
18 CDE: Concept-Driven Exploration for Reinforcement Learning 提出概念驱动探索(CDE)方法，解决视觉强化学习中高效探索问题 manipulation reinforcement learning
19 Towards Reliable LLM-based Robot Planning via Combined Uncertainty Estimation 提出CURE，结合不确定性估计，提升LLM机器人规划的可靠性 manipulation distillation large language model
20 BLAZER: Bootstrapping LLM-based Manipulation Agents with Zero-Shot Data Generation BLAZER：利用零样本数据生成引导基于LLM的机器人操作代理 manipulation
21 EB-MBD: Emerging-Barrier Model-Based Diffusion for Safe Trajectory Optimization in Highly Constrained Environments 提出基于新兴障碍函数的模型扩散方法，用于高约束环境下的安全轨迹优化。 trajectory optimization

🔬 支柱二：RL算法与架构 (RL & Architecture) (3 篇)

#题目一句话要点标签🔗⭐
22 Geometry-aware Policy Imitation 提出几何感知策略模仿(GPI)，解决机器人模仿学习中效率和泛化性问题。 imitation learning multimodal ✅
23 Learning to Navigate Socially Through Proactive Risk Perception 提出基于主动风险感知的社交导航方法，提升动态人群环境下的机器人导航能力 privileged information egocentric
24 Evaluation of a Robust Control System in Real-World Cable-Driven Parallel Robots TRPO在欠驱动缆索驱动并联机器人控制中表现出卓越的鲁棒性 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🔗⭐
25 NavSpace: How Navigation Agents Follow Spatial Intelligence Instructions 提出NavSpace基准测试，评估并提升导航Agent的空间智能 large language model multimodal instruction following
26 ConPoSe: LLM-Guided Contact Point Selection for Scalable Cooperative Object Pushing ConPoSe：基于LLM引导的接触点选择，实现可扩展的协作物体推移 large language model

⬅️ 返回 cs.RO 首页 · 🏠 返回主页