cs.RO（2026-02-20）

📊 共 15 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (9 🔗2) 支柱三：空间感知与语义 (Perception & Semantics) (2) 支柱九：具身大模型 (Embodied Foundation Models) (2 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (2)

🔬 支柱一：机器人控制 (Robot Control) (9 篇)

#题目一句话要点标签🔗⭐
1 Soft Surfaced Vision-Based Tactile Sensing for Bipedal Robot Applications 提出基于视觉的软表面触觉足传感器，提升双足机器人平衡与环境感知能力 legged robot legged locomotion bipedal
2 SimVLA: A Simple VLA Baseline for Robotic Manipulation SimVLA：用于机器人操作的简单且强大的视觉-语言-动作基线模型 manipulation vision-language-action VLA ✅
3 GrandTour: A Legged Robotics Dataset in the Wild for Multi-Modal Perception and State Estimation GrandTour：用于多模态感知和状态估计的野外足式机器人数据集 quadruped legged robot ANYmal
4 EgoPush: Learning End-to-End Egocentric Multi-Object Rearrangement for Mobile Robots EgoPush：学习基于自中心视觉的多物体重排列策略，用于移动机器人 sim-to-real reinforcement learning policy learning ✅
5 Learning to Tune Pure Pursuit in Autonomous Racing: Joint Lookahead and Steering-Gain Control with PPO 提出基于PPO的Pure Pursuit参数自整定方法，提升自主赛车性能 MPC reinforcement learning PPO
6 Learning Smooth Time-Varying Linear Policies with an Action Jacobian Penalty 提出基于动作雅可比惩罚的线性策略网络，用于学习平滑的时变线性策略。 quadruped parkour reinforcement learning
7 Role-Adaptive Collaborative Formation Planning for Team of Quadruped Robots in Cluttered Environments 提出一种角色自适应的四足机器人集群编队规划方法，解决复杂环境下的协同导航问题。 quadruped
8 Snapping Actuators with Asymmetric and Sequenced Motion 提出基于几何诱导不稳定性的非对称吸动器，实现可控软体机器人运动 quadruped locomotion
9 Zero-shot Interactive Perception 提出零样本交互感知框架，结合多策略操作与记忆驱动的视觉语言模型，解决复杂场景中的语义查询问题。 manipulation affordance

🔬 支柱三：空间感知与语义 (Perception & Semantics) (2 篇)

#题目一句话要点标签🔗⭐
10 OVerSeeC: Open-Vocabulary Costmap Generation from Satellite Images and Natural Language OVerSeeC：基于卫星图像和自然语言的开放词汇代价地图生成 open-vocabulary open vocabulary foundation model
11 Have We Mastered Scale in Deep Monocular Visual SLAM? The ScaleMaster Dataset and Benchmark 提出ScaleMaster数据集，揭示深度单目SLAM在尺度一致性上的不足 visual SLAM

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

#题目一句话要点标签🔗⭐
12 How Fast Can I Run My VLA? Demystifying VLA Inference Performance with VLA-Perf VLA-Perf：用于分析和优化Vision-Language-Action模型推理性能的分析模型 embodied AI vision-language-action VLA
13 RoEL: Robust Event-based 3D Line Reconstruction 提出RoEL方法以解决事件相机3D线重建问题 multimodal ✅

🔬 支柱二：RL算法与架构 (RL & Architecture) (2 篇)

#题目一句话要点标签🔗⭐
14 Toward AI Autonomous Navigation for Mechanical Thrombectomy using Hierarchical Modular Multi-agent Reinforcement Learning (HM-MARL) 提出HM-MARL框架，实现机械取栓术中导管导丝的自主导航 reinforcement learning world model
15 Interacting safely with cyclists using Hamilton-Jacobi reachability and reinforcement learning 提出基于Hamilton-Jacobi可达性与强化学习的框架，保障自动驾驶车辆与骑行者交互安全 reinforcement learning

⬅️ 返回 cs.RO 首页 · 🏠 返回主页