cs.RO（2026-02-10）

📊 共 33 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (26 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (3) 支柱二：RL算法与架构 (RL & Architecture) (3 🔗1) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱一：机器人控制 (Robot Control) (26 篇)

#题目一句话要点标签🔗⭐
1 EgoHumanoid: Unlocking In-the-Wild Loco-Manipulation with Robot-Free Egocentric Demonstration EgoHumanoid：利用无机器人的人类自我中心演示，实现野外环境的人形机器人操作 humanoid humanoid control manipulation
2 VLA-JEPA: Enhancing Vision-Language-Action Model with Latent World Model VLA-JEPA：利用潜在世界模型增强视觉-语言-动作模型，提升泛化性和鲁棒性 manipulation world model vision-language-action
3 TeleGate: Whole-Body Humanoid Teleoperation via Gated Expert Selection with Motion Prior TeleGate：基于门控专家选择与运动先验的全身人形机器人遥操作 humanoid humanoid robot fall recovery
4 DexImit: Learning Bimanual Dexterous Manipulation from Monocular Human Videos DexImit：从单目人类视频学习灵巧双手动手操作 manipulation dexterous hand dexterous manipulation
5 BagelVLA: Enhancing Long-Horizon Manipulation via Interleaved Vision-Language-Action Generation BagelVLA：通过交错式视觉-语言-动作生成增强长时程操作能力 manipulation vision-language-action VLA
6 CAPER: Constrained and Procedural Reasoning for Robotic Scientific Experiments CAPER：约束与程序推理框架，提升机器人科学实验的可靠性与数据效率 manipulation reinforcement learning policy learning
7 Phase-Aware Policy Learning for Skateboard Riding of Quadruped Robots via Feature-wise Linear Modulation 提出基于相位感知的强化学习方法PAPL，用于四足机器人滑板控制 quadruped legged robot locomotion
8 Diverse Skill Discovery for Quadruped Robots via Unsupervised Learning 提出正交混合专家与多判别器框架，提升四足机器人无监督技能发现的多样性与效率 quadruped locomotion Unitree
9 Sci-VLA: Agentic VLA Inference Plugin for Long-Horizon Tasks in Scientific Experiments 提出Agentic VLA插件，解决科学实验中长时程任务的VLA模型推理难题。 manipulation vision-language-action VLA
10 UniVTAC: A Unified Simulation Platform for Visuo-Tactile Manipulation Data Generation, Learning, and Benchmarking UniVTAC：用于视觉-触觉操作数据生成、学习和评估的统一仿真平台 manipulation policy learning vision-language-action ✅
11 Humanoid Factors: Design Principles for AI Humanoids in Human Worlds 提出人形因素框架以解决人形机器人设计挑战 humanoid humanoid robot humanoid control
12 RoboSubtaskNet: Temporal Sub-task Segmentation for Human-to-Robot Skill Transfer in Real-World Environments 提出RoboSubtaskNet，用于真实环境中人机协作技能迁移的子任务分割 manipulation optical flow human-to-robot
13 TriPilot-FF: Coordinated Whole-Body Teleoperation with Force Feedback TriPilot-FF：力反馈协同双臂移动机器人全身遥操作系统 manipulation bi-manual teleoperation
14 RoboInter: A Holistic Intermediate Representation Suite Towards Robotic Manipulation RoboInter：面向机器人操作的整体中间表示套件，提升VLA模型泛化性。 manipulation vision-language-action VLA
15 Preference Aligned Visuomotor Diffusion Policies for Deformable Object Manipulation 提出RKO方法，利用少量演示对预训练扩散策略进行偏好对齐，提升机器人对可变形物体操作的个性化能力。 manipulation diffusion policy preference learning
16 Robo3R: Enhancing Robotic Manipulation with Accurate Feed-Forward 3D Reconstruction Robo3R：通过精确的前馈3D重建增强机器人操作 manipulation sim-to-real motion planning
17 Decoupled MPPI-Based Multi-Arm Motion Planning 提出基于解耦MPPI的多臂机器人运动规划算法MR-STORM MPC model predictive control motion planning
18 Learning Force-Regulated Manipulation with a Low-Cost Tactile-Force-Controlled Gripper 提出RETAF框架和低成本力觉触觉gripper，实现机器人精确力控操作 manipulation teleoperation policy learning
19 AnyTouch 2: General Optical Tactile Representation Learning For Dynamic Tactile Perception AnyTouch 2：面向动态触觉感知的通用光学触觉表征学习框架 manipulation dexterous manipulation representation learning
20 A Collision-Free Sway Damping Model Predictive Controller for Safe and Reactive Forestry Crane Navigation 提出一种无碰撞摆动阻尼模型预测控制器以解决林业起重机导航问题 MPC model predictive control
21 Certified Gradient-Based Contact-Rich Manipulation via Smoothing-Error Reachable Tubes 提出基于平滑误差可达管的梯度优化方法，用于保证接触丰富的操作任务。 manipulation dexterous manipulation
22 Sample-Efficient Real-World Dexterous Policy Fine-Tuning via Action-Chunked Critics and Normalizing Flows SOFT-FLOW：基于Normalizing Flow和分块Critic的真实世界灵巧策略高效微调 manipulation dexterous manipulation multimodal
23 Instruct2Act: From Human Instruction to Actions Sequencing and Execution via Robot Action Network for Robotic Manipulation Instruct2Act：通过机器人动作网络实现从人类指令到机器人操作的动作序列生成与执行 manipulation
24 Fast Motion Planning for Non-Holonomic Mobile Robots via a Rectangular Corridor Representation of Structured Environments 提出基于矩形走廊表示的快速非完整移动机器人运动规划方法 motion planning
25 Learning Agile Quadrotor Flight in the Real World 提出自适应框架，解决四旋翼在真实世界中敏捷飞行的难题。 sim2real
26 Optimal Control of Microswimmers for Trajectory Tracking Using Bayesian Optimization 提出基于贝叶斯优化的微型游泳器轨迹跟踪最优控制方法 locomotion

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

#题目一句话要点标签🔗⭐
27 AutoFly: Vision-Language-Action Model for UAV Autonomous Navigation in the Wild AutoFly：面向野外无人机自主导航的视觉-语言-动作模型 embodied AI vision-language-action VLA
28 Design and Evaluation of an Assisted Programming Interface for Behavior Trees in Robotics BETR-GUI：结合AI辅助的行为树编程界面，提升机器人程序开发效率 large language model
29 LLM-Grounded Dynamic Task Planning with Hierarchical Temporal Logic for Human-Aware Multi-Robot Collaboration 提出基于LLM和分层时序逻辑的多机器人人机协作动态任务规划方法 large language model

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

#题目一句话要点标签🔗⭐
30 ST4VLA: Spatially Guided Training for Vision-Language-Action Models ST4VLA：利用空间引导训练提升视觉-语言-动作模型在具身任务中的性能 policy learning vision-language-action VLA ✅
31 Rethinking Visual-Language-Action Model Scaling: Alignment, Mixture, and Regularization 重新思考视觉-语言-动作模型扩展：对齐、混合与正则化 flow matching cross-embodiment embodiment transfer
32 NavDreamer: Video Models as Zero-Shot 3D Navigators NavDreamer：利用视频模型实现零样本3D导航 dreamer spatiotemporal vision-language-action

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

#题目一句话要点标签🔗⭐
33 TaCo: A Benchmark for Lossless and Lossy Codecs of Heterogeneous Tactile Data TaCo：异构触觉数据编解码基准，提升机器人实时触觉感知能力 spatiotemporal

⬅️ 返回 cs.RO 首页 · 🏠 返回主页