cs.RO（2026-04-22）

📊 共 17 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (11 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (3) 支柱九：具身大模型 (Embodied Foundation Models) (2) 支柱五：交互与反应 (Interaction & Reaction) (1)

🔬 支柱一：机器人控制 (Robot Control) (11 篇)

#题目一句话要点标签🔗⭐
1 VTouch++: A Multimodal Dataset with Vision-Based Tactile Enhancement for Bimanual Manipulation VTouch++：用于双臂操作的视觉触觉增强多模态数据集 manipulation bi-manual bimanual manipulation
2 PokeVLA: Empowering Pocket-Sized Vision-Language-Action Model with Comprehensive World Knowledge Guidance PokeVLA：融合世界知识的轻量级视觉-语言-动作模型，赋能口袋机器人操作 manipulation affordance vision-language-action ✅
3 JoyAI-RA 0.1: A Foundation Model for Robotic Autonomy 提出JoyAI-RA，一个用于机器人自主的视觉-语言-动作具身基础模型 manipulation egocentric cross-embodiment
4 Cortex 2.0: Grounding World Models in Real-World Industrial Deployment Cortex 2.0：基于世界模型的工业机器人长期操作规划与部署 manipulation dual-arm world model
5 FingerEye: Continuous and Unified Vision-Tactile Sensing for Dexterous Manipulation FingerEye：用于灵巧操作的连续统一视觉-触觉传感 manipulation dexterous manipulation imitation learning ✅
6 Bimanual Robot Manipulation via Multi-Agent In-Context Learning BiCICLe：基于多智能体上下文学习的双臂机器人操作框架 manipulation bi-manual bimanual manipulation
7 ETac: A Lightweight and Efficient Tactile Simulation Framework for Learning Dexterous Manipulation ETac：一种轻量高效的触觉模拟框架，用于灵巧操作学习。 manipulation dexterous manipulation reinforcement learning
8 Passive Variable Impedance For Shared Control 提出一种被动可变阻抗共享控制方法，用于稳定机器人操作臂的阻抗控制。 shared control
9 Lexicographic Minimum-Violation Motion Planning using Signal Temporal Logic 提出基于信号时序逻辑的最小违例运动规划方法，解决自动驾驶中多重冲突规范的优先级满足问题。 motion planning
10 Visual-Tactile Peg-in-Hole Assembly Learning from Peg-out-of-Hole Disassembly 提出一种视觉-触觉融合的PiH学习框架，利用PooH逆向任务提升效率。 manipulation reinforcement learning policy learning
11 Kinematic Optimization of Phalanx Length Ratios in Robotic Hands Using Potential Dexterity 提出基于潜在灵巧度的机器人手掌指节长度比例优化方法 manipulation

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

#题目一句话要点标签🔗⭐
12 Temporal Difference Calibration in Sequential Tasks: Application to Vision-Language-Action Models 提出基于时序差分的校准方法，提升视觉-语言-动作模型在序列任务中的不确定性估计。 reinforcement learning vision-language-action VLA
13 Toward Safe Autonomous Robotic Endovascular Interventions using World Models 提出基于世界模型的TD-MPC2算法，实现安全自主的机器人血管内介入 reinforcement learning SAC world model
14 AdaTracker: Learning Adaptive In-Context Policy for Cross-Embodiment Active Visual Tracking AdaTracker：学习自适应上下文策略，实现跨形态机器人主动视觉跟踪 policy learning cross-embodiment

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

#题目一句话要点标签🔗⭐
15 A Vision-Language-Action Model for Adaptive Ultrasound-Guided Needle Insertion and Needle Tracking 提出VLA模型，用于超声引导下自适应针头插入与跟踪 vision-language-action VLA
16 LLM-Guided Safety Agent for Edge Robotics with an ISO-Compliant Perception-Compute-Control Architecture 提出LLM引导的安全代理，用于符合ISO标准的边缘机器人人机交互安全 embodied AI

🔬 支柱五：交互与反应 (Interaction & Reaction) (1 篇)

#题目一句话要点标签🔗⭐
17 ALAS: Adaptive Long-Horizon Action Synthesis via Async-pathway Stream Disentanglement ALAS：基于异步路径流解耦的自适应长时程动作合成，提升人机交互任务性能。 human-scene interaction HSI spatial relationship

⬅️ 返回 cs.RO 首页 · 🏠 返回主页