cs.RO（2025-05-28）

📊 共 21 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (12 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (3 🔗1) 支柱三：空间感知与语义 (Perception & Semantics) (3 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (2 🔗1) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱一：机器人控制 (Robot Control) (12 篇)

#题目一句话要点标签🔗⭐
1 ForceVLA: Enhancing VLA Models with a Force-aware MoE for Contact-rich Manipulation ForceVLA：力觉感知的混合专家模型增强接触式操作的VLA模型 manipulation dexterous manipulation vision-language-action
2 From Motion to Behavior: Hierarchical Modeling of Humanoid Generative Behavior Control 提出GBC框架，利用LLM驱动的分层行为计划生成多样且逼真的人形行为 humanoid motion planning motion generation
3 FastTD3: Simple, Fast, and Capable Reinforcement Learning for Humanoid Control FastTD3：一种简单、快速且适用于人形机器人控制的强化学习算法 humanoid humanoid robot humanoid control
4 DexUMI: Using Human Hand as the Universal Manipulation Interface for Dexterous Manipulation DexUMI：利用人手作为通用操控界面，实现灵巧操作技能迁移 manipulation dexterous manipulation policy learning
5 ReinFlow: Fine-tuning Flow Matching Policy with Online Reinforcement Learning ReinFlow：在线强化学习微调Flow Matching策略，提升连续机器人控制性能 legged locomotion locomotion manipulation
6 TwinTrack: Bridging Vision and Contact Physics for Real-Time Tracking of Unknown Dynamic Objects TwinTrack：融合视觉与接触物理，实现未知动态物体实时跟踪 manipulation in-hand manipulation sim2real
7 Learning Compositional Behaviors from Demonstration and Language BLADE：融合语言与示教的长程机器人操作行为学习框架 manipulation imitation learning large language model
8 Prior Reinforce: Mastering Agile Tasks with Limited Trials Prior Reinforce：通过少量试验掌握敏捷任务的强化学习方法 manipulation motion planning reward design ✅
9 Spring-Brake! Handed Shearing Auxetics Improve Efficiency of Hopping and Standing 利用手性剪切膨胀结构，提升单足跳跃机器人跳跃和站立的能效 legged robot locomotion
10 LabUtopia: High-Fidelity Simulation and Hierarchical Benchmark for Scientific Embodied Agents LabUtopia：为科学具身智能体提供高保真模拟与分层基准测试。 manipulation mobile manipulation
11 VR-Based Control of Multi-Copter Operation 提出基于VR的无人机遥操作系统，提升未知环境下的安全性。 teleoperation
12 Spot-On: A Mixed Reality Interface for Multi-Robot Cooperation 提出一种混合现实界面，用于多足机器人协作完成复杂任务。 quadruped

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

#题目一句话要点标签🔗⭐
13 ChatVLA-2: Vision-Language-Action Model with Open-World Embodied Reasoning from Pretrained Knowledge ChatVLA-2：基于预训练知识的开放世界具身推理视觉-语言-动作模型 imitation learning vision-language-action VLA
14 Streaming Flow Policy: Simplifying diffusion/flow-matching policies by treating action trajectories as flow trajectories 提出流式Flow策略，通过将动作轨迹视为流轨迹简化扩散/Flow匹配策略 imitation learning flow matching
15 SCIZOR: A Self-Supervised Approach to Data Curation for Large-Scale Imitation Learning SCIZOR：一种自监督数据清洗方法，用于提升大规模模仿学习性能。 imitation learning ✅

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

#题目一句话要点标签🔗⭐
16 UP-SLAM: Adaptively Structured Gaussian SLAM with Uncertainty Prediction in Dynamic Environments UP-SLAM：动态环境下基于不确定性预测的自适应结构高斯SLAM 3D gaussian splatting 3DGS gaussian splatting ✅
17 DORAEMON: Decentralized Ontology-aware Reliable Agent with Enhanced Memory Oriented Navigation DORAEMON：面向家庭服务机器人的去中心化、本体感知、可靠增强记忆导航Agent scene understanding spatiotemporal
18 LiDAR Based Semantic Perception for Forklifts in Outdoor Environments 提出一种基于双激光雷达的语义分割框架，用于室外叉车自主导航。 scene understanding

🔬 支柱九：具身大模型 (Embodied Foundation Models) (2 篇)

#题目一句话要点标签🔗⭐
19 Dynamic Task Adaptation for Multi-Robot Manufacturing Systems with Large Language Models 提出基于大语言模型的多机器人制造系统动态任务自适应框架 large language model
20 Anomalies by Synthesis: Anomaly Detection using Generative Diffusion Models for Off-Road Navigation 提出基于生成扩散模型的异常检测方法，用于越野导航中识别未知环境障碍物。 foundation model ✅

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

#题目一句话要点标签🔗⭐
21 COSMOS: A Data-Driven Probabilistic Time Series simulator for Chemical Plumes across Spatial Scales COSMOS：一种数据驱动的概率时间序列模拟器，用于模拟跨空间尺度的化学羽流 spatiotemporal

⬅️ 返回 cs.RO 首页 · 🏠 返回主页