cs.RO（2026-03-02）

📊 共 27 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (18 🔗3) 支柱九：具身大模型 (Embodied Foundation Models) (4 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (2) 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) (18 篇)

#题目一句话要点标签🔗⭐
1 PhysGraph: Physically-Grounded Graph-Transformer Policies for Bimanual Dexterous Hand-Tool-Object Manipulation 提出PhysGraph，用于解决灵巧手-工具-物体双手动灵巧操作中的高维状态空间问题。 manipulation dexterous hand dexterous manipulation
2 Learning Thermal-Aware Locomotion Policies for an Electrically-Actuated Quadruped Robot 提出热感知强化学习方法，提升电驱动四足机器人续航能力 quadruped locomotion locomotion policy
3 TacMamba: A Tactile History Compression Adapter Bridging Fast Reflexes and Slow VLA Reasoning TacMamba：触觉历史压缩适配器，连接快速触觉反射与慢速视觉推理 manipulation Mamba representation learning
4 Mean-Flow based One-Step Vision-Language-Action 提出基于均值流的单步视觉-语言-动作方法，加速机器人操作任务中的动作生成。 manipulation diffusion policy flow matching
5 ACDC: Adaptive Curriculum Planning with Dynamic Contrastive Control for Goal-Conditioned Reinforcement Learning in Robotic Manipulation ACDC：结合动态对比控制的自适应课程规划，提升机器人操作中的目标条件强化学习性能 manipulation reinforcement learning contrastive learning
6 Tiny-DroNeRF: Tiny Neural Radiance Fields aboard Federated Learning-enabled Nano-drones Tiny-DroNeRF：在算力受限的纳米无人机上实现联邦学习NeRF，用于复杂环境三维重建。 motion planning NeRF neural radiance field
7 Bimanual XR Specification of Relative and Absolute Assembly Hierarchies for Teleoperation 提出一种基于双手动XR交互的遥操作装配层级结构规范方法 bi-manual teleoperation
8 Non-Markovian Long-Horizon Robot Manipulation via Keyframe Chaining 提出Keyframe-Chaining VLA模型，解决长时程非马尔可夫机器人操作任务 manipulation vision-language-action VLA ✅
9 Jailbreaking Embodied LLMs via Action-level Manipulation 提出Blindfold框架，通过动作级操控破解具身LLM的安全防护 manipulation embodied AI large language model
10 Rethinking Camera Choice: An Empirical Study on Fisheye Camera Properties in Robotic Manipulation 针对机器人操作，论文深入研究鱼眼相机特性，并提出提升策略。 manipulation policy learning imitation learning
11 SFCo-Nav: Efficient Zero-Shot Visual Language Navigation via Collaboration of Slow LLM and Fast Attributed Graph Alignment SFCo-Nav：通过慢速LLM与快速属性图对齐协同，实现高效的零样本视觉语言导航 legged robot VLN large language model
12 Closed-Loop Action Chunks with Dynamic Corrections for Training-Free Diffusion Policy DCDP：动态闭环扩散策略，无需重训练即可提升动态环境下的机器人操作适应性。 manipulation diffusion policy ✅
13 From Transportation to Manipulation: Transforming Magnetic Levitation to Magnetic Robotics 提出基于磁悬浮的六自由度机器人平台，实现运输到操作的转变 manipulation
14 A Safety-Aware Shared Autonomy Framework with BarrierIK Using Control Barrier Functions 提出基于控制障碍函数的安全共享自主框架以解决环境拥挤问题 teleoperation VR teleoperation ✅
15 Towards Robot Skill Learning and Adaptation with Gaussian Processes 提出基于高斯过程的机器人技能学习与自适应框架，提升复杂任务泛化能力。 manipulation reinforcement learning
16 CHOP: Counterfactual Human Preference Labels Improve Obstacle Avoidance in Visuomotor Navigation Policies CHOP：利用反事实人类偏好标签提升视觉运动导航策略的避障能力 quadruped
17 SaferPath: Hierarchical Visual Navigation with Learned Guidance and Safety-Constrained Control SaferPath：结合学习引导与安全约束控制的分层视觉导航框架 MPC
18 Shape-Interpretable Visual Self-Modeling Enables Geometry-Aware Continuum Robot Control 提出基于形状可解释视觉自建模的连续体机器人几何感知控制方法 manipulation

🔬 支柱九：具身大模型 (Embodied Foundation Models) (4 篇)

#题目一句话要点标签🔗⭐
19 Neural Implicit Action Fields: From Discrete Waypoints to Continuous Functions for Vision-Language-Action Models 提出神经隐式动作场（NIAF），将VLA模型的离散路径点预测转为连续动作函数回归。 vision-language-action VLA
20 KERV: Kinematic-Rectified Speculative Decoding for Embodied VLA Models 提出KERV：一种结合运动学矫正的推测解码框架，加速具身VLA模型推理。 vision-language-action VLA
21 Multimodal Adversarial Quality Policy for Safe Grasping 提出多模态对抗质量策略MAQP，保障RGBD机器人抓取的安全性 multimodal
22 LAD-Drive: Bridging Language and Trajectory with Action-Aware Diffusion Transformers LAD-Drive：利用动作感知扩散Transformer桥接语言与轨迹，提升自动驾驶决策能力。 large language model multimodal ✅

🔬 支柱二：RL算法与架构 (RL & Architecture) (2 篇)

#题目一句话要点标签🔗⭐
23 Learning Vision-Based Omnidirectional Navigation: A Teacher-Student Approach Using Monocular Depth Estimation 提出基于单目深度估计的师生学习框架，实现全向移动机器人的视觉导航。 PPO teacher-student depth estimation
24 $π$-StepNFT: Wider Space Needs Finer Steps in Online RL for Flow-based VLAs 提出π-StepNFT，解决基于流的VLA模型在线强化学习中多步采样难题。 reinforcement learning vision-language-action VLA

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

#题目一句话要点标签🔗⭐
25 SSMG-Nav: Enhancing Lifelong Object Navigation with Semantic Skeleton Memory Graph 提出SSMG-Nav，利用语义骨架记忆图增强终身物体导航能力 traversability multimodal
26 FATE: Closed-Loop Feasibility-Aware Task Generation with Active Repair for Physically Grounded Robotic Curricula FATE：通过主动修复的闭环可行性感知任务生成，用于物理机器人课程学习 affordance large language model

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

#题目一句话要点标签🔗⭐
27 Real-Time Thermal-Inertial Odometry on Embedded Hardware for High-Speed GPS-Denied Flight 提出一种实时热惯性里程计，用于嵌入式硬件上的高速、无GPS环境飞行。 PULSE

⬅️ 返回 cs.RO 首页 · 🏠 返回主页