cs.RO（2025-12-17）
📊 共 18 篇论文 | 🔗 1 篇有代码
🎯 兴趣领域导航
支柱一：机器人控制 (Robot Control) (6)
支柱三：空间感知与语义 (Perception & Semantics) (4)
支柱九：具身大模型 (Embodied Foundation Models) (3)
支柱二：RL算法与架构 (RL & Architecture) (2)
支柱七：动作重定向 (Motion Retargeting) (1)
支柱六：视频提取与匹配 (Video Extraction) (1)
支柱八：物理动画 (Physics-based Animation) (1 🔗1)
🔬 支柱一：机器人控制 (Robot Control) (6 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 1 | mimic-video: Video-Action Models for Generalizable Robot Control Beyond VLAs | 提出mimic-video，一种基于视频的动作模型，提升机器人控制的泛化性和样本效率。 | manipulation flow matching vision-language-action | ||
| 2 | ISS Policy : Scalable Diffusion Policy with Implicit Scene Supervision | 提出基于隐式场景监督的可扩展扩散策略，提升机器人操作任务的泛化性和训练效率。 | manipulation dexterous hand imitation learning | ||
| 3 | GuangMing-Explorer: A Four-Legged Robot Platform for Autonomous Exploration in General Environments | GuangMing-Explorer：用于通用环境自主探索的四足机器人平台 | legged robot | ||
| 4 | SORS: A Modular, High-Fidelity Simulator for Soft Robots | SORS：用于软体机器人的模块化、高保真模拟器，提升仿真到现实的迁移能力。 | sim-to-real | ||
| 5 | QuantGraph: A Receding-Horizon Quantum Graph Solver | 提出QuantGraph，一种基于后退视界的量子图求解器，提升图优化效率。 | model predictive control | ||
| 6 | Load-Based Variable Transmission Mechanism for Robotic Applications | 提出基于负载的可变传动机制，无需额外执行器即可动态调整机器人关节扭矩。 | legged robot |
🔬 支柱三：空间感知与语义 (Perception & Semantics) (4 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 7 | VLA-AN: An Efficient and Onboard Vision-Language-Action Framework for Aerial Navigation in Complex Environments | VLA-AN：用于复杂环境无人机自主导航的高效、端侧视觉-语言-动作框架 | 3D gaussian splatting gaussian splatting splatting | ||
| 8 | OMCL: Open-vocabulary Monte Carlo Localization | 提出基于视觉-语言特征的开放词汇蒙特卡洛定位方法，提升跨模态地图环境下的机器人定位鲁棒性。 | open-vocabulary open vocabulary | ||
| 9 | NAP3D: NeRF Assisted 3D-3D Pose Alignment for Autonomous Vehicles | NAP3D：NeRF辅助的3D-3D位姿对齐，用于提升自动驾驶车辆定位精度 | NeRF neural radiance field geometric consistency | ||
| 10 | HERO: Hierarchical Traversable 3D Scene Graphs for Embodied Navigation Among Movable Obstacles | HERO：用于可移动障碍物环境具身导航的分层可遍历3D场景图 | traversability |
🔬 支柱九：具身大模型 (Embodied Foundation Models) (3 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 11 | Large Video Planner Enables Generalizable Robot Control | 提出基于大规模视频预训练的通用机器人控制框架，实现零样本泛化。 | vision-language-action VLA large language model | ||
| 12 | MiVLA: Towards Generalizable Vision-Language-Action Model with Human-Robot Mutual Imitation Pre-training | MiVLA：通过人-机互模仿预训练实现通用视觉-语言-动作模型 | vision-language-action VLA | ||
| 13 | Trust in LLM-controlled Robotics: a Survey of Security Threats, Defenses and Challenges | 综述LLM控制机器人中的安全威胁与防御，为安全可靠的机器人系统提供蓝图 | large language model |
🔬 支柱二：RL算法与架构 (RL & Architecture) (2 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 14 | Modeling the Mental World for Embodied AI: A Comprehensive Review | 构建具身AI心智模型：提出完整理论框架，促进人机协作 | world model embodied AI | ||
| 15 | SWIFT-Nav: Stability-Aware Waypoint-Level TD3 with Fuzzy Arbitration for UAV Navigation in Cluttered Environments | 提出SWIFT-Nav，结合模糊仲裁与TD3，提升UAV在复杂环境导航的稳定性和效率。 | TD3 reward shaping |
🔬 支柱七：动作重定向 (Motion Retargeting) (1 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 16 | Few-Shot Inference of Human Perceptions of Robot Performance in Social Navigation Scenarios | 利用少量样本，通过大语言模型预测社交导航中人类对机器人性能的感知 | human motion large language model |
🔬 支柱六：视频提取与匹配 (Video Extraction) (1 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 17 | BEV-Patch-PF: Particle Filtering with BEV-Aerial Feature Matching for Off-Road Geo-Localization | 提出BEV-Patch-PF，利用BEV特征匹配的粒子滤波实现越野环境无GPS定位 | feature matching |
🔬 支柱八：物理动画 (Physics-based Animation) (1 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 18 | dLITE: Differentiable Lighting-Informed Trajectory Evaluation for On-Orbit Inspection | 提出dLITE以解决轨道检查中图像质量优化问题 | differentiable simulation | ✅ |