cs.RO（2024-10-15）

📊 共 18 篇论文 | 🔗 6 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (14 🔗3) 支柱三：空间感知与语义 (Perception & Semantics) (2 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (1 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (1 🔗1)

🔬 支柱一：机器人控制 (Robot Control) (14 篇)

#题目一句话要点标签🔗⭐
1 Dynamic Open-Vocabulary 3D Scene Graphs for Long-term Language-Guided Mobile Manipulation DovSG：用于长期语言引导移动操作的动态开放词汇3D场景图 manipulation mobile manipulation scene reconstruction ✅
2 Learning Smooth Humanoid Locomotion through Lipschitz-Constrained Policies 提出Lipschitz约束策略，实现人形机器人平滑鲁棒的强化学习运动控制 legged robot humanoid humanoid robot
3 M2Diffuser: Diffusion-based Trajectory Optimization for Mobile Manipulation in 3D Scenes M2Diffuser：基于扩散模型的3D场景移动操作轨迹优化 locomotion manipulation mobile manipulation
4 OKAMI: Teaching Humanoid Robots Manipulation Skills through Single Video Imitation OKAMI：通过单视频模仿教人形机器人操作技能 humanoid humanoid robot manipulation ✅
5 Adaptive Ankle Torque Control for Bipedal Humanoid Walking on Surfaces with Unknown Horizontal and Vertical Motion 提出自适应踝关节力矩控制，解决双足人形机器人在未知运动表面上的稳定行走问题 humanoid bipedal biped
6 Visual Manipulation with Legs 提出基于视觉的四足机器人腿部操作系统，实现非抓取操作 quadruped locomotion manipulation
7 PAVLM: Advancing Point Cloud based Affordance Understanding Via Vision-Language Model PAVLM：利用视觉语言模型提升点云的Affordance理解 manipulation affordance large language model
8 Latent Action Pretraining from Videos 提出LAPA：一种基于视频潜在动作预训练的通用动作模型，无需人工标注。 manipulation VQ-VAE vision-language-action
9 Learning Goal-oriented Bimanual Dough Rolling Using Dynamic Heterogeneous Graph Based on Human Demonstration 提出基于动态异构图的模仿学习方法，用于解决软物料双臂擀面任务 humanoid humanoid robot manipulation
10 SDS -- See it, Do it, Sorted: Quadruped Skill Synthesis from Single Video Demonstration SDS：从单视频演示中合成四足机器人技能，无需标签或奖励工程。 quadruped locomotion Unitree ✅
11 DeformPAM: Data-Efficient Learning for Long-horizon Deformable Object Manipulation via Preference-based Action Alignment 提出DeformPAM，通过偏好学习高效解决长时程柔性物体操作任务 manipulation imitation learning preference learning
12 Generalizable Spacecraft Trajectory Generation via Multimodal Learning with Transformers 提出基于Transformer多模态学习的通用航天器轨迹生成框架，提升复杂场景适应性。 trajectory optimization multimodal
13 Robust Manipulation Primitive Learning via Domain Contraction 提出基于领域收缩的鲁棒操作原语学习方法，提升接触式操作的泛化性。 manipulation domain randomization policy learning
14 Towards Local Minima-free Robotic Navigation: Model Predictive Path Integral Control via Repulsive Potential Augmentation 提出基于斥力势场增强的MPPI控制，解决机器人导航局部极小值问题 motion planning

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

#题目一句话要点标签🔗⭐
15 GSORB-SLAM: Gaussian Splatting SLAM benefits from ORB features and Transmittance information GSORB-SLAM：融合ORB特征与透射率信息的3D高斯溅射SLAM，提升跟踪精度与重建质量。 visual SLAM 3D gaussian splatting 3DGS ✅
16 NavTopo: Leveraging Topological Maps For Autonomous Navigation Of a Mobile Robot NavTopo：利用拓扑地图实现移动机器人的自主导航 occupancy grid

🔬 支柱二：RL算法与架构 (RL & Architecture) (1 篇)

#题目一句话要点标签🔗⭐
17 Learning from 10 Demos: Generalisable and Sample-Efficient Policy Learning with Oriented Affordance Frames 提出基于Oriented Affordance Frames的策略学习方法，仅需少量演示即可泛化 policy learning imitation learning affordance ✅

🔬 支柱九：具身大模型 (Embodied Foundation Models) (1 篇)

#题目一句话要点标签🔗⭐
18 LLM2Swarm: Robot Swarms that Responsively Reason, Plan, and Collaborate through LLMs LLM2Swarm：利用LLM实现机器人集群的响应式推理、规划与协作 large language model ✅

⬅️ 返回 cs.RO 首页 · 🏠 返回主页