cs.RO（2025-12-22）

📊 共 20 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (12) 支柱二：RL算法与架构 (RL & Architecture) (3) 支柱九：具身大模型 (Embodied Foundation Models) (3 🔗1) 支柱七：动作重定向 (Motion Retargeting) (2 🔗1)

🔬 支柱一：机器人控制 (Robot Control) (12 篇)

#题目一句话要点标签🔗⭐
1 Affordance RAG: Hierarchical Multimodal Retrieval with Affordance-Aware Embodied Memory for Mobile Manipulation Affordance RAG：用于移动操作的具身记忆分层多模态检索 manipulation mobile manipulation open-vocabulary
2 A Framework for Deploying Learning-based Quadruped Loco-Manipulation 提出基于强化学习的四足机器人灵巧操作部署框架，解决仿真到现实迁移难题 quadruped whole-body control manipulation
3 TwinAligner: Visual-Dynamic Alignment Empowers Physics-aware Real2Sim2Real for Robotic Manipulation TwinAligner：视觉-动力学对齐赋能物理感知Real2Sim2Real机器人操作 manipulation sim2real real2sim
4 EGM: Efficiently Learning General Motion Tracking Policy for High Dynamic Humanoid Whole-Body Control EGM：高效学习通用运动跟踪策略，用于高动态人形机器人全身控制 humanoid whole-body control human motion
5 REALM: A Real-to-Sim Validated Benchmark for Generalization in Robotic Manipulation REALM：一个经验证的真实-模拟机器人操作泛化基准 manipulation vision-language-action VLA
6 A Flexible Field-Based Policy Learning Framework for Diverse Robotic Systems and Sensors 提出基于场信息的柔性策略学习框架，实现跨机器人和传感器的操作技能泛化 manipulation bi-manual teleoperation
7 Learning Generalizable Hand-Object Tracking from Synthetic Demonstrations 提出HOP+HOT框架，仅用合成数据学习通用手-物跟踪控制器 manipulation dexterous hand dexterous manipulation
8 LeLaR: The First In-Orbit Demonstration of an AI-Based Satellite Attitude Controller LeLaR：首次在轨演示基于AI的卫星姿态控制器，解决传统控制器的设计难题。 sim2real reinforcement learning deep reinforcement learning
9 Results of the 2024 CommonRoad Motion Planning Competition for Autonomous Vehicles CommonRoad自动驾驶运动规划竞赛：标准化评估复杂交通场景下的规划算法 motion planning
10 Translating Flow to Policy via Hindsight Online Imitation 提出HinFlow，通过回溯在线模仿学习将高层规划转化为机器人策略 manipulation cross-embodiment
11 Real2Edit2Real: Generating Robotic Demonstrations via a 3D Control Interface Real2Edit2Real：通过3D控制界面生成机器人操作演示，提升数据效率。 manipulation
12 Collaborative Continuum Robots: A Survey 综述协作式连续体机器人，提升任务适应性、工作空间和负载能力。 motion planning

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

#题目一句话要点标签🔗⭐
13 CoDrone: Autonomous Drone Navigation Assisted by Edge and Cloud Foundation Models CoDrone：边缘云协同，利用基础模型增强无人机自主导航能力 reinforcement learning deep reinforcement learning depth estimation
14 WorldRFT: Latent World Model Planning with Reinforcement Fine-Tuning for Autonomous Driving WorldRFT：通过强化微调的潜在世界模型规划，提升自动驾驶安全性。 reinforcement learning world model representation learning
15 DTCCL: Disengagement-Triggered Contrastive Continual Learning for Autonomous Bus Planners 提出DTCCL框架，通过Disengagement触发的对比持续学习提升自动驾驶巴士规划策略 imitation learning contrastive learning

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

#题目一句话要点标签🔗⭐
16 IndoorUAV: Benchmarking Vision-Language UAV Navigation in Continuous Indoor Environments 提出IndoorUAV以解决室内无人机视觉语言导航问题 embodied AI VLA VLN
17 PalpAid: Multimodal Pneumatic Tactile Sensor for Tissue Palpation 提出PalpAid以解决机器人手术中的触觉信息缺失问题 multimodal
18 MaP-AVR: A Meta-Action Planner for Agents Leveraging Vision Language Models and Retrieval-Augmented Generation 提出MaP-AVR，利用VLM和RAG的机器人元动作规划器 chain-of-thought ✅

🔬 支柱七：动作重定向 (Motion Retargeting) (2 篇)

#题目一句话要点标签🔗⭐
19 LoGoPlanner: Localization Grounded Navigation Policy with Metric-aware Visual Geometry LoGoPlanner：基于度量视觉几何的定位引导端到端导航策略 cross-embodiment ✅
20 Vision-Language-Policy Model for Dynamic Robot Task Planning 提出VLP模型，解决机器人动态任务规划中指令理解与策略更新难题 cross-embodiment

⬅️ 返回 cs.RO 首页 · 🏠 返回主页