cs.RO（2026-02-04）

📊 共 20 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (13 🔗4) 支柱三：空间感知与语义 (Perception & Semantics) (3) 支柱九：具身大模型 (Embodied Foundation Models) (3) 支柱六：视频提取与匹配 (Video Extraction) (1)

🔬 支柱一：机器人控制 (Robot Control) (13 篇)

#	题目	一句话要点	标签	🔗	⭐
1	PDF-HR: Pose Distance Fields for Humanoid Robots	提出PDF-HR：基于位姿距离场的人形机器人运动先验模型	humanoid humanoid robot reward shaping
2	GeneralVLA: Generalizable Vision-Language-Action Models with Knowledge-Guided Trajectory Planning	提出 GeneralVLA，通过知识引导轨迹规划实现机器人零样本操作	manipulation behavior cloning affordance	✅
3	HoRD: Robust Humanoid Control via History-Conditioned Reinforcement Learning and Online Distillation	提出HoRD框架，通过历史条件强化学习和在线蒸馏实现鲁棒的人形机器人控制。	humanoid humanoid robot humanoid control	✅
4	A Unified Complementarity-based Approach for Rigid-Body Manipulation and Motion Prediction	提出Unicomp：一种统一的基于互补性的刚体操作与运动预测方法	manipulation predictive model motion prediction
5	EgoActor: Grounding Task Planning into Spatial-aware Egocentric Actions for Humanoid Robots via Visual-Language Models	EgoActor：通过视觉-语言模型将任务规划融入空间感知的以自我为中心的具身智能动作	humanoid humanoid robot locomotion
6	Act, Sense, Act: Learning Non-Markovian Active Perception Strategies from Large-Scale Egocentric Human Data	提出CoMe-VLA框架，利用大规模人类数据学习非马尔可夫主动感知策略，提升机器人操作能力。	humanoid manipulation egocentric
7	Reshaping Action Error Distributions for Reliable Vision-Language-Action Models	针对连续动作VLA模型，提出基于最小误差熵的训练方法，提升泛化性和鲁棒性	manipulation vision-language-action VLA	✅
8	Viewpoint Matters: Dynamically Optimizing Viewpoints with Masked Autoencoder for Visual Manipulation	提出MAE-Select，利用掩码自编码器动态优化机器人操作的视角选择。	manipulation imitation learning masked autoencoder
9	GeoLanG: Geometry-Aware Language-Guided Grasping with Unified RGB-D Multimodal Learning	GeoLanG：提出几何感知语言引导抓取框架，解决复杂场景下的多模态机器人操作问题。	manipulation multimodal
10	ALORE: Autonomous Large-Object Rearrangement with a Legged Manipulator	ALORE：基于腿式机器人的自主大型物体重排系统，提升复杂环境下的操作效率。	whole-body control manipulation loco-manipulation	✅
11	Integrated Exploration and Sequential Manipulation on Scene Graph with LLM-based Situated Replanning	EPoG：基于LLM情境重规划的场景图探索与序列操作集成框架	manipulation large language model
12	Capturing Visual Environment Structure Correlates with Control Performance	通过环境状态解码评估视觉表征，提升机器人控制策略泛化性	manipulation
13	Shaping Expressiveness in Robotics: The Role of Design Tools in Crafting Embodied Robot Movements	提出一种基于设计工具的机器人表达性运动生成方法，提升人机交互体验	manipulation

🔬 支柱三：空间感知与语义 (Perception & Semantics) (3 篇)

#	题目	一句话要点	标签	🔗	⭐
14	Towards Next-Generation SLAM: A Survey on 3DGS-SLAM Focusing on Performance, Robustness, and Future Directions	综述3DGS-SLAM：面向性能、鲁棒性和未来方向的下一代SLAM系统	3D gaussian splatting 3DGS gaussian splatting
15	Relational Scene Graphs for Object Grounding of Natural Language Commands	提出基于关系场景图的目标物体定位方法，提升LLM在人机交互中的指令理解能力。	open-vocabulary open vocabulary large language model
16	TACO: Temporal Consensus Optimization for Continual Neural Mapping	提出TACO，通过时序共识优化实现持续神经地图构建，解决动态环境中适应性问题。	scene understanding

🔬 支柱九：具身大模型 (Embodied Foundation Models) (3 篇)

#	题目	一句话要点	标签	🔗	⭐
17	SCALE: Self-uncertainty Conditioned Adaptive Looking and Execution for Vision-Language-Action Models	提出SCALE，通过自不确定性调节视觉感知和动作执行，提升VLA模型鲁棒性。	vision-language-action VLA
18	A Modern System Recipe for Situated Embodied Human-Robot Conversation with Real-Time Multimodal LLMs and Tool-Calling	提出基于实时多模态LLM和工具调用的具身人机对话系统	large language model multimodal
19	The Supportiveness-Safety Tradeoff in LLM Well-Being Agents	研究LLM在心理健康支持Agent中支持性和安全性之间的权衡	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	🔗	⭐
20	From Vision to Assistance: Gaze and Vision-Enabled Adaptive Control for a Back-Support Exoskeleton	提出基于视觉和眼动追踪的自适应控制方法，提升腰部外骨骼的辅助效果。	egocentric egocentric vision

⬅️ 返回 cs.RO 首页 · 🏠 返回主页