cs.RO（2026-03-12）

📊 共 28 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (25 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (1) 支柱三：空间感知与语义 (Perception & Semantics) (1) 支柱九：具身大模型 (Embodied Foundation Models) (1)

🔬 支柱一：机器人控制 (Robot Control) (25 篇)

#题目一句话要点标签🔗⭐
1 $Ψ_0$: An Open Foundation Model Towards Universal Humanoid Loco-Manipulation 提出Ψ₀，一个开放的通用人形机器人Loco-Manipulation基础模型 humanoid humanoid robot manipulation
2 HumDex:Humanoid Dexterous Manipulation Made Easy HumDex：一种便捷的人形机器人灵巧操作遥操作与学习系统 humanoid manipulation dexterous hand ✅
3 SPARK: Skeleton-Parameter Aligned Retargeting on Humanoid Robots with Kinodynamic Trajectory Optimization SPARK：基于骨骼参数对齐和运动动力学轨迹优化的人形机器人动作重定向 humanoid humanoid robot humanoid control
4 Sim-to-reality adaptation for Deep Reinforcement Learning applied to an underwater docking application 提出基于深度强化学习的水下机器人自主对接Sim-to-Real迁移方法 sim-to-real reinforcement learning deep reinforcement learning
5 SaPaVe: Towards Active Perception and Manipulation in Vision-Language-Action Models for Robotics 提出SaPaVe以解决机器人主动感知与操作统一问题 manipulation vision-language-action ✅
6 ComFree-Sim: A GPU-Parallelized Analytical Contact Physics Engine for Scalable Contact-Rich Robotics Simulation and Control ComFree-Sim：一种GPU并行化的解析接触物理引擎，用于可扩展的富接触机器人仿真与控制。 manipulation dexterous manipulation MPC
7 CoViLLM: An Adaptive Human-Robot Collaborative Assembly Framework Using Large Language Models for Manufacturing 提出CoViLLM框架以解决定制化产品组装的灵活性问题 manipulation large language model
8 Energy Prediction on Sloping Ground for Quadruped Robots 针对四足机器人，提出基于坡度信息的地面能量预测模型 quadruped legged robot locomotion
9 MiNI-Q: A Miniature, Wire-Free Quadruped with Unbounded, Independently Actuated Leg Joints MiNI-Q^2：一种具有无界独立驱动腿部关节的微型无线四足机器人 quadruped legged robot locomotion
10 Concurrent Prehensile and Nonprehensile Manipulation: A Practical Approach to Multi-Stage Dexterous Tasks DexMulti：一种用于灵巧手多阶段操作的、高效的技能复用框架 manipulation dexterous hand diffusion policy
11 RoboClaw: An Agentic Framework for Scalable Long-Horizon Robotic Tasks RoboClaw：面向可扩展长时程机器人任务的Agentic框架 manipulation policy learning vision-language-action
12 HandelBot: Real-World Piano Playing via Fast Adaptation of Dexterous Robot Policies HandelBot：通过快速自适应灵巧机器人策略实现真实钢琴演奏 manipulation dexterous manipulation bi-manual
13 CRAFT: A Tendon-Driven Hand with Hybrid Hard-Soft Compliance CRAFT：一种混合软硬度、肌腱驱动的灵巧手，适用于接触丰富的操作任务 manipulation teleoperation
14 Diversity You Can Actually Measure: A Fast, Model-Free Diversity Metric for Robotics Datasets 提出FAKTUAL，一种快速、无模型的机器人数据集多样性度量与数据筛选方法，提升模仿学习泛化性能。 manipulation imitation learning
15 Coupling Tensor Trains with Graph of Convex Sets: Effective Compression, Exploration, and Planning in the C-Space 提出TANGO，结合张量网络与凸集图优化，实现高效构型空间压缩、探索与规划。 motion planning
16 A Hybrid Neural-Assisted Unscented Kalman Filter for Unmanned Ground Vehicle Navigation 提出一种混合神经辅助的Unscented Kalman滤波器，用于提升无人地面车辆导航精度。 sim2real
17 A Generalized Theory of Load Distribution in Redundantly-actuated Robotic Systems 提出一种广义理论，解决冗余驱动机器人系统中的载荷分布问题 legged robot
18 Grounding Robot Generalization in Training Data via Retrieval-Augmented VLMs RADAR：利用检索增强的视觉语言模型，将机器人泛化能力与训练数据对齐 manipulation
19 HumDex: Humanoid Dexterous Manipulation Made Easy HumDex：一种便捷的人形机器人灵巧操作系统，提升数据收集效率。 humanoid manipulation dexterous hand ✅
20 Beyond Motion Imitation: Is Human Motion Data Alone Sufficient to Explain Gait Control and Biomechanics? 基于强化学习的运动模仿需考虑足地交互，以保证步态控制的生物力学合理性 gait control reinforcement learning imitation learning
21 COAD: Constant-Time Planning for Continuous Goal Manipulation with Compressed Library and Online Adaptation COAD：基于压缩库和在线自适应的连续目标操作恒定时间规划 manipulation trajectory optimization motion planning ✅
22 Push, Press, Slide: Mode-Aware Planar Contact Manipulation via Reduced-Order Models 提出一种基于降阶模型的模式感知平面接触操作框架，用于单臂和双臂机器人操作。 manipulation bi-manual bimanual manipulation
23 One-Step Flow Policy: Self-Distillation for Fast Visuomotor Policies 提出One-Step Flow Policy，通过自蒸馏实现快速高精度机器人视觉运动策略。 manipulation distillation multimodal
24 GNN-DIP: Neural Corridor Selection for Decomposition-Based Motion Planning GNN-DIP：基于图神经网络的分解运动规划走廊选择方法 motion planning
25 Robots that redesign themselves through kinematic self-destruction 提出一种基于运动学自毁的自设计机器人，通过自主重塑优化运动性能。 locomotion

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

#题目一句话要点标签🔗⭐
26 NFPO: Stabilized Policy Optimization of Normalizing Flow for Robotic Policy Learning 提出NFPO：一种稳定的Normalizing Flow策略优化方法，用于机器人策略学习。 reinforcement learning deep reinforcement learning DRL

🔬 支柱三：空间感知与语义 (Perception & Semantics) (1 篇)

#题目一句话要点标签🔗⭐
27 HiSync: Spatio-Temporally Aligning Hand Motion from Wearable IMU and On-Robot Camera for Command Source Identification in Long-Range HRI HiSync：通过光惯性融合实现远距离人机交互中的命令源识别 optical flow multimodal

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

#题目一句话要点标签🔗⭐
28 Red-Teaming Vision-Language-Action Models via Quality Diversity Prompt Generation for Robust Robot Policies 提出Q-DIG，通过质量多样性提示生成对抗样本，提升VLA机器人策略的鲁棒性 vision-language-action VLA

⬅️ 返回 cs.RO 首页 · 🏠 返回主页