cs.RO（2025-12-30）

📊 共 12 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (8 🔗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) (1) 支柱二：RL算法与架构 (RL & Architecture) (1)

🔬 支柱一：机器人控制 (Robot Control) (8 篇)

#题目一句话要点标签🔗⭐
1 GR-Dexter Technical Report GR-Dexter：面向灵巧手双臂机器人的通用操作VLA框架 manipulation dexterous hand bi-manual
2 Subsecond 3D Mesh Generation for Robot Manipulation 提出一种亚秒级3D网格生成系统，用于机器人操作中的实时场景感知。 manipulation open-vocabulary open vocabulary
3 World In Your Hands: A Large-Scale and Open-source Ecosystem for Learning Human-centric Manipulation in the Wild 提出WiYH大规模开放生态系统，用于学习以人为中心的野外操作 manipulation dexterous hand policy learning
4 Unified Embodied VLM Reasoning with Robotic Action via Autoregressive Discretized Pre-training 提出GenieReasoner，通过自回归离散预训练统一具身VLM推理与机器人动作 manipulation flow matching vision-language-action ✅
5 Evaluation of Impression Difference of a Domestic Mobile Manipulator with Autonomous and/or Remote Control in Fetch-and-Carry Tasks 评估自主与遥控模式下移动操作机器人在取放任务中的用户体验差异 manipulation mobile manipulation teleoperation
6 Real-world Reinforcement Learning from Suboptimal Interventions 提出SiLRI算法，利用次优干预加速真实机器人操作强化学习 manipulation teleoperation reinforcement learning ✅
7 3D Path-Following Guidance via Nonlinear Model Predictive Control for Fixed-Wing Small UAS 针对固定翼小型无人机，提出基于非线性模型预测控制的3D路径跟踪制导算法 MPC model predictive control
8 Local Path Optimization in The Latent Space Using Learned Distance Gradient 提出基于学习距离梯度的局部路径优化方法，加速机器人约束运动规划。 manipulation 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🔗⭐
9 Foundation models on the bridge: Semantic hazard detection and safety maneuvers for maritime autonomy with vision-language models 提出Semantic Lookout，利用视觉-语言模型实现海上自主航行的语义危险检测与安全规避。 scene understanding IMoS foundation model
10 RANGER: A Monocular Zero-Shot Semantic Navigation Framework through Contextual Adaptation RANGER：基于上下文自适应的单目零样本语义导航框架 open-vocabulary open vocabulary foundation model

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

#题目一句话要点标签🔗⭐
11 Counterfactual VLA: Self-Reflective Vision-Language-Action Model with Adaptive Reasoning 提出CF-VLA，通过反事实推理提升自动驾驶决策安全性和准确性。 vision-language-action VLA

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

#题目一句话要点标签🔗⭐
12 DRL-TH: Jointly Utilizing Temporal Graph Attention and Hierarchical Fusion for UGV Navigation in Crowded Environments 提出DRL-TH框架，利用时序图注意力与分层融合提升UGV在拥挤环境中的导航能力 reinforcement learning deep reinforcement learning DRL

⬅️ 返回 cs.RO 首页 · 🏠 返回主页