cs.RO（2026-03-13）

📊 共 24 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (15 🔗2) 支柱九：具身大模型 (Embodied Foundation Models) (3) 支柱二：RL算法与架构 (RL & Architecture) (3 🔗1) 支柱三：空间感知与语义 (Perception & Semantics) (2) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱一：机器人控制 (Robot Control) (15 篇)

#题目一句话要点标签🔗⭐
1 TacVLA: Contact-Aware Tactile Fusion for Robust Vision-Language-Action Manipulation TacVLA：融合触觉信息的视觉-语言-动作操作模型，提升机器人操作鲁棒性 manipulation contact-aware vision-language-action
2 Altered Thoughts, Altered Actions: Probing Chain-of-Thought Vulnerabilities in VLA Robotic Manipulation VLA机器人操作中思维链的脆弱性研究：中间推理过程的对抗攻击 manipulation vision-language-action VLA
3 Learning Athletic Humanoid Tennis Skills from Imperfect Human Motion Data LATENT：从不完美人类动作数据学习类人机器人网球技能 humanoid humanoid robot sim-to-real ✅
4 Beyond Dense Futures: World Models as Structured Planners for Robotic Manipulation StructVLA：通过结构化规划提升机器人操作的世界模型 manipulation world model spatiotemporal
5 AnchorVLA4D: an Anchor-Based Spatial-Temporal Vision-Language-Action Model for Robotic Manipulation AnchorVLA4D：基于锚点的时空视觉-语言-动作机器人操作模型 manipulation vision-language-action VLA
6 FLUX: Accelerating Cross-Embodiment Generative Navigation Policies via Rectified Flow and Static-to-Dynamic Learning FLUX：通过修正流和静态到动态学习加速跨具身生成式导航策略 quadruped humanoid sim-to-real
7 Panoramic Multimodal Semantic Occupancy Prediction for Quadruped Robots 针对四足机器人，提出PanoMMOcc数据集和VoxelHound框架，实现全景多模态语义占据预测。 quadruped multimodal ✅
8 Coordinated Manipulation of Hybrid Deformable-Rigid Objects in Constrained Environments 提出基于优化的混合柔性-刚性物体协同操作规划器，解决约束环境下的操作难题。 manipulation dual-arm trajectory optimization
9 SmoothTurn: Learning to Turn Smoothly for Agile Navigation with Quadrupedal Robots SmoothTurn：学习平滑转向，助力四足机器人敏捷导航 quadruped locomotion locomotion policy
10 Beyond Imitation: Reinforcement Learning Fine-Tuning for Adaptive Diffusion Navigation Policies 提出基于强化学习微调的自适应扩散导航策略，提升机器人泛化能力。 quadruped reinforcement learning imitation learning
11 Language-Grounded Decoupled Action Representation for Robotic Manipulation 提出LaDA框架，通过解耦动作表示和语义引导学习，提升机器人操作的泛化性和动作一致性。 manipulation contrastive learning curriculum learning
12 MotionAnymesh: Physics-Grounded Articulation for Simulation-Ready Digital Twins MotionAnymesh：提出基于物理的关节运动框架，为仿真环境构建即用数字孪生 trajectory optimization penetration embodied AI
13 Evaluating VLMs' Spatial Reasoning Over Robot Motion: A Step Towards Robot Planning with Motion Preferences 评估VLMs在机器人运动中的空间推理能力，助力具备运动偏好的机器人规划 motion planning foundation model
14 RoboStream: Weaving Spatio-Temporal Reasoning with Memory in Vision-Language Models for Robotics RoboStream：融合时空推理与记忆的视觉-语言模型，提升机器人操作能力 manipulation VoxPoser
15 From Woofs to Words: Towards Intelligent Robotic Guide Dogs with Verbal Communication 为导盲犬机器人设计口语交流系统，提升人机协作决策能力 quadruped

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

#题目一句话要点标签🔗⭐
16 ReMem-VLA: Empowering Vision-Language-Action Model with Memory via Dual-Level Recurrent Queries ReMem-VLA：通过双层循环查询增强视觉-语言-动作模型的记忆能力 vision-language-action VLA OpenVLA
17 SldprtNet: A Large-Scale Multimodal Dataset for CAD Generation in Language-Driven 3D Design SldprtNet：用于语言驱动3D设计中CAD生成的大规模多模态数据集 multimodal
18 DecoVLN: Decoupling Observation, Reasoning, and Correction for Vision-and-Language Navigation DecoVLN：解耦观察、推理和纠正，用于视觉语言导航 VLN

🔬 支柱二：RL算法与架构 (RL & Architecture) (3 篇)

#题目一句话要点标签🔗⭐
19 Efficient Real-World Autonomous Racing via Attenuated Residual Policy Optimization 提出衰减残差策略优化算法，实现高效的真实世界自主赛车 reinforcement learning deep reinforcement learning DRL ✅
20 Reinforcement Learning for Elliptical Cylinder Motion Control Tasks 提出基于强化学习的椭圆柱体运动控制方法，解决受限扭矩下的控制难题 reinforcement learning
21 Easy-IIL: Reducing Human Operational Burden in Interactive Imitation Learning via Assistant Experts Easy-IIL：利用辅助专家降低交互式模仿学习中的人工操作负担 imitation learning

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

#题目一句话要点标签🔗⭐
22 GoalSwarm: Multi-UAV Semantic Coordination for Open-Vocabulary Object Navigation 提出GoalSwarm以解决多无人机开放词汇目标导航问题 open-vocabulary open vocabulary foundation model
23 HaltNav: Reactive Visual Halting over Lightweight Topological Priors for Robust Vision-Language Navigation HaltNav：基于轻量级拓扑先验的反应式视觉停止导航，提升视觉语言导航的鲁棒性 open-vocabulary open vocabulary VLN

🔬 支柱八：物理动画 (Physics-based Animation) (1 篇)

#题目一句话要点标签🔗⭐
24 AoI-FusionNet: Age-Aware Tightly Coupled Fusion of UWB-IMU under Sparse Ranging Conditions 提出AoI-FusionNet，解决稀疏测距下UWB-IMU紧耦合融合的雪崩粒子追踪问题 motion tracking

⬅️ 返回 cs.RO 首页 · 🏠 返回主页