cs.RO（2026-03-17）

📊 共 37 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (27 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (5) 支柱三：空间感知与语义 (Perception & Semantics) (3 🔗1) 支柱七：动作重定向 (Motion Retargeting) (1) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱一：机器人控制 (Robot Control) (27 篇)

#题目一句话要点标签🔗⭐
1 Rewarding DINO: Predicting Dense Rewards with Vision Foundation Models Rewarding DINO：利用视觉基础模型预测稠密奖励，提升机器人操作任务性能 manipulation reinforcement learning foundation model
2 OGScene3D: Incremental Open-Vocabulary 3D Gaussian Scene Graph Mapping for Scene Understanding 提出OGScene3D以解决增量开放词汇3D场景理解问题 manipulation scene understanding semantic mapping
3 Towards the Vision-Sound-Language-Action Paradigm: The HEAR Framework for Sound-Centric Manipulation 提出HEAR框架以解决声音中心操控中的实时性问题 manipulation flow matching world model
4 Enabling Dynamic Tracking in Vision-Language-Action Models via Time-Discrete and Time-Continuous Velocity Feedforward 通过时序离散与连续的速度前馈，增强视觉-语言-动作模型中的动态跟踪能力 manipulation teleoperation behavior cloning
5 DexGrasp-Zero: A Morphology-Aligned Policy for Zero-Shot Cross-Embodiment Dexterous Grasping 提出DexGrasp-Zero，通过形态对齐策略实现灵巧手抓取的零样本跨具身迁移。 dexterous hand policy learning cross-embodiment
6 Learning Whole-Body Control for a Salamander Robot 提出基于强化学习的蝾螈机器人全身控制方法，实现水陆环境下的稳定运动。 quadruped legged robot whole-body control
7 Enforcing Task-Specified Compliance Bounds for Humanoids via Anisotropic Lipschitz-Constrained Policies 提出各向异性Lipschitz约束策略以解决人形机器人合规性问题 humanoid humanoid robot bipedal
8 MolmoB0T: Large-Scale Simulation Enables Zero-Shot Manipulation MolmoBot：大规模仿真实现机器人零样本操作 manipulation mobile manipulation sim-to-real
9 Beyond Cybathlon: On-demand Quadrupedal Assistance for People with Limited Mobility 提出基于共享自主的四足机器人辅助系统，提升行动不便人士的独立性 quadruped manipulation mobile manipulation
10 ADAPT: Adaptive Dual-projection Architecture for Perceptive Traversal 提出ADAPT，通过自适应双投影架构实现复杂3D环境中的高效拟人运动 humanoid humanoid locomotion locomotion
11 Onboard MuJoCo-based Model Predictive Control for Shipboard Crane with Double-Pendulum Sway Suppression 提出基于MuJoCo MPC的船载起重机双摆抑制方法，解决海上扰动下的重物转移难题。 MPC model predictive control reinforcement learning
12 Dexterous grasp data augmentation based on grasp synthesis with fingertip workspace cloud and contact-aware sampling 提出基于指尖工作空间云和接触感知采样的灵巧抓取数据增强方法 teleoperation contact-aware
13 DreamPlan: Efficient Reinforcement Fine-Tuning of Vision-Language Planners via Video World Models DreamPlan：通过视频世界模型高效强化微调视觉-语言规划器 manipulation reinforcement learning world model
14 Large Reward Models: Generalizable Online Robot Reward Generation with Vision-Language Models 提出基于视觉-语言模型的在线机器人奖励生成框架，提升强化学习策略泛化性。 manipulation reinforcement learning imitation learning
15 The Era of End-to-End Autonomy: Transitioning from Rule-Based Driving to Large Driving Models 从规则到学习：自动驾驶迈向端到端大模型的时代 humanoid humanoid robot embodied AI
16 Encoding Predictability and Legibility for Style-Conditioned Diffusion Policy 提出风格条件扩散策略(SCDP)，在人机协作中平衡运动效率和可理解性。 manipulation diffusion policy
17 CABTO: Context-Aware Behavior Tree Grounding for Robot Manipulation CABTO：基于上下文感知的行为树自动构建框架，用于机器人操作 manipulation
18 A Pin-Array Structured Climbing Robot for Stable Locomotion on Steep Rocky Terrain 提出一种基于柔顺针阵列结构的攀爬机器人，用于在陡峭岩石地形上稳定运动 locomotion
19 LIMBERO: A Limbed Climbing Exploration Robot Toward Traveling on Rocky Cliffs LIMBERO：一种用于在岩石峭壁上行进的四足攀爬机器人 quadruped legged robot locomotion
20 MG-Grasp: Metric-Scale Geometric 6-DoF Grasping Framework with Sparse RGB Observations MG-Grasp：基于稀疏RGB图像的度量尺度几何6自由度抓取框架 manipulation foundation model
21 Geometry-Aligned LLM Fine-Tuning for Sequential Narrow-Opening Planning 提出几何对齐的LLM微调框架，用于解决序列窄口运动规划问题 motion planning large language model
22 ManiTwin: Scaling Data-Generation-Ready Digital Object Dataset to 100K ManiTwin：构建大规模可用于数据生成的十万级数字物体数据集 manipulation policy learning ✅
23 Kamino: GPU-based Massively Parallel Simulation of Multi-Body Systems with Challenging Topologies Kamino：基于GPU的大规模并行多体系统仿真，支持复杂拓扑结构 biped reinforcement learning
24 BrickSim: A Physics-Based Simulator for Manipulating Interlocking Brick Assemblies BrickSim：用于操作互锁积木组件的基于物理的实时仿真器 manipulation ✅
25 vAccSOL: Efficient and Transparent AI Vision Offloading for Mobile Robots vAccSOL：用于移动机器人的高效透明AI视觉卸载框架 quadruped
26 When Rolling Gets Weird: A Curved-Link Tensegrity Robot for Non-Intuitive Behavior 提出一种基于曲杆张拉整体结构的机器人，用于实现非直观运动。 locomotion
27 Featurized Occupation Measures for Structured Global Search in Numerical Optimal Control 提出特征化占用测度(FOM)，用于数值最优控制中的结构化全局搜索。 trajectory optimization

🔬 支柱二：RL算法与架构 (RL & Architecture) (5 篇)

#题目一句话要点标签🔗⭐
28 Kinema4D: Kinematic 4D World Modeling for Spatiotemporal Embodied Simulation Kinema4D：用于时空具身仿真的运动学4D世界建模 world model physically plausible spatiotemporal
29 When Should a Robot Think? Resource-Aware Reasoning via Reinforcement Learning for Embodied Robotic Decision-Making RARRL：基于强化学习的资源感知推理框架，提升具身机器人决策效率 reinforcement learning large language model
30 Conservative Offline Robot Policy Learning via Posterior-Transition Reweighting 提出PTR方法，通过后验-转移重加权实现异构机器人离线策略学习。 policy learning flow matching
31 Agile Interception of a Flying Target using Competitive Reinforcement Learning 提出基于竞争强化学习的无人机敏捷拦截方法，解决复杂动态环境下的目标追踪问题。 reinforcement learning PPO
32 Controlling Fish Schools via Reinforcement Learning of Virtual Fish Movement 提出基于强化学习的虚拟鱼运动控制方法，引导鱼群行为 reinforcement learning

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

#题目一句话要点标签🔗⭐
33 SE(3)-LIO: Smooth IMU Propagation With Jointly Distributed Poses on SE(3) Manifold for Accurate and Robust LiDAR-Inertial Odometry 提出SE(3)-LIO，通过SE(3)流形上的平滑IMU传播实现精确鲁棒的激光雷达惯性里程计 LIO motion prediction ✅
34 GenZ-LIO: Generalizable LiDAR-Inertial Odometry Beyond Indoor--Outdoor Boundaries GenZ-LIO：提出一种可泛化室内外环境的激光雷达惯性里程计框架 LIO
35 Industrial cuVSLAM Benchmark & Integration 针对工业物流环境，提出并验证了基于cuVSLAM的移动机器人视觉SLAM基准测试与集成方案。 visual odometry visual SLAM

🔬 支柱七：动作重定向 (Motion Retargeting) (1 篇)

#题目一句话要点标签🔗⭐
36 PanguMotion: Continuous Driving Motion Forecasting with Pangu Transformers PanguMotion：利用Pangu Transformer实现连续驾驶场景的运动轨迹预测 motion prediction large language model

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

#题目一句话要点标签🔗⭐
37 Thermopneumatic Pixels for Fast, Localized, Low-Voltage Touch Feedback 提出热 пневматический 像素，实现快速、局部、低压触觉反馈 PULSE

⬅️ 返回 cs.RO 首页 · 🏠 返回主页