cs.RO（2025-03-10）

📊 共 34 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (21) 支柱九：具身大模型 (Embodied Foundation Models) (6 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (4) 支柱三：空间感知与语义 (Perception & Semantics) (2 🔗1) 支柱四：生成式动作 (Generative Motion) (1)

🔬 支柱一：机器人控制 (Robot Control) (21 篇)

#题目一句话要点标签🔗⭐
1 VMTS: Vision-Assisted Teacher-Student Reinforcement Learning for Multi-Terrain Locomotion in Bipedal Robots 提出基于视觉辅助的师生强化学习方法，提升双足机器人多地形适应性 legged robot bipedal biped
2 A Task and Motion Planning Framework Using Iteratively Deepened AND/OR Graph Networks 提出一种基于迭代加深的AND/OR图网络的任务与运动规划框架，解决子任务数量未知的TAMP问题。 dual-arm motion planning task and motion planning
3 Physics-informed Neural Network Predictive Control for Quadruped Locomotion 提出基于在线负载识别的物理信息神经网络预测控制，解决四足机器人负载未知下的精确运动控制问题 quadruped legged locomotion locomotion
4 RoboCopilot: Human-in-the-loop Interactive Imitation Learning for Robot Manipulation RoboCopilot：用于机器人操作的人在环交互式模仿学习 manipulation bi-manual teleoperation
5 Unlocking Generalization for Robotics via Modularity and Scale 结合模块化与大规模学习，提升机器人泛化能力以实现通用机器人控制 manipulation sim2real policy learning
6 VidBot: Learning Generalizable 3D Actions from In-the-Wild 2D Human Videos for Zero-Shot Robotic Manipulation VidBot：利用野外2D人体视频学习通用3D动作，实现零样本机器人操作 manipulation affordance foundation model
7 One-Shot Dual-Arm Imitation Learning 提出One-Shot双臂模仿学习(ODIL)，仅需单次演示即可学习精确协同的双臂任务。 dual-arm imitation learning
8 A2I-Calib: An Anti-noise Active Multi-IMU Spatial-temporal Calibration Framework for Legged Robots 提出A2I-Calib框架，解决腿足机器人多IMU系统时空标定中的噪声敏感问题。 quadruped legged robot reinforcement learning
9 iManip: Skill-Incremental Learning for Robotic Manipulation 提出iManip框架，解决机器人操作技能增量学习中的灾难性遗忘问题。 manipulation generalist agent
10 How to Train Your Robots? The Impact of Demonstration Modality on Imitation Learning 研究示教方式对模仿学习的影响，提出混合示教策略提升机器人学习性能 manipulation teleoperation imitation learning
11 Multi-layer Motion Planning with Kinodynamic and Spatio-Temporal Constraints 提出多层运动规划方法，解决车辆在时空和运动约束下的路径生成问题 motion planning spatiotemporal
12 Delta-Triplane Transformers as Occupancy World Models 提出Delta-Triplane Transformer，用于高效预测自动驾驶场景中的占用世界模型 motion planning world model
13 Focused Blind Switching Manipulation Based on Constrained and Regional Touch States of Multi-Fingered Hand Using Deep Learning 提出基于深度学习的约束触觉状态盲切换操作方法，用于灵巧手盖子开启任务 manipulation
14 Force Aware Branch Manipulation To Assist Agricultural Tasks 提出力感知的分枝操作方法，辅助农业机器人执行任务 manipulation
15 WHERE-Bot: a Wheel-less Helical-ring Everting Robot Capable of Omnidirectional Locomotion 提出WHERE-Bot：一种无轮螺旋环外翻机器人，实现全向运动，适用于复杂地形探索。 locomotion
16 LTLCodeGen: Code Generation of Syntactically Correct Temporal Logic for Robot Task Planning LTLCodeGen：用于机器人任务规划的句法正确时序逻辑代码生成 motion planning large language model
17 A Decapod Robot with Rotary Bellows-Enclosed Soft Transmissions 提出一种基于旋转波纹管软传输的十足软体机器人，实现高效地形适应性运动。 legged robot locomotion
18 Geometric Retargeting: A Principled, Ultrafast Neural Hand Retargeting Algorithm 提出Geometric Retargeting (GeoRT)，一种超快速、有原则的神经手部重定向算法，用于遥操作。 teleoperation
19 QBIT: Quality-Aware Cloud-Based Benchmarking for Robotic Insertion Tasks QBIT：面向机器人插入任务的质量感知云基准测试框架 sim-to-real
20 Multi-Robot System for Cooperative Exploration in Unknown Environments: A Survey 多机器人协同探索未知环境综述：模块化框架与未来挑战 motion planning
21 Eye-in-Finger: Smart Fingers for Delicate Assembly and Disassembly of LEGO 提出Eye-in-Finger工具设计，解决乐高积木精细装配中视觉遮挡问题 manipulation

🔬 支柱九：具身大模型 (Embodied Foundation Models) (6 篇)

#题目一句话要点标签🔗⭐
22 SafePlan: Leveraging Formal Logic and Chain-of-Thought Reasoning for Enhanced Safety in LLM-based Robotic Task Planning SafePlan：结合形式逻辑与思维链推理，提升LLM机器人任务规划安全性 large language model chain-of-thought
23 Bi-Directional Mental Model Reconciliation for Human-Robot Interaction with Large Language Models 提出基于大语言模型的双向心智模型协调框架，提升人机交互效果 large language model
24 Self-Corrective Task Planning by Inverse Prompting with Large Language Models 提出InversePrompt，利用逆向提示实现LLM机器人任务规划的自校正 large language model
25 Safety Guardrails for LLM-Enabled Robots RoboGuard：为LLM赋能的机器人设计安全防护栏，应对幻觉和对抗攻击。 large language model chain-of-thought
26 HELM: Human-Preferred Exploration with Language Models 提出HELM：利用语言模型实现人类偏好引导的自主探索 large language model
27 Graphormer-Guided Task Planning: Beyond Static Rules with LLM Safety Perception 提出Graphormer引导的任务规划框架，提升LLM在机器人安全任务中的风险感知和适应性。 large language model ✅

🔬 支柱二：RL算法与架构 (RL & Architecture) (4 篇)

#题目一句话要点标签🔗⭐
28 PointVLA: Injecting the 3D World into Vision-Language-Action Models PointVLA：将3D世界注入视觉-语言-动作模型，无需重训练。 imitation learning diffusion policy vision-language-action
29 Towards Safe Robot Foundation Models 提出基于ATACOM的安全层，提升机器人通用策略在现实场景中的安全性 reinforcement learning foundation model
30 AffordDexGrasp: Open-set Language-guided Dexterous Grasp with Generalizable-Instructive Affordance 提出AffordDexGrasp框架，解决开放集语言引导的灵巧抓取问题。 flow matching affordance
31 Research and Design on Intelligent Recognition of Unordered Targets for Robots Based on Reinforcement Learning 提出基于强化学习的智能机器人无序目标识别方法，提升复杂场景下的识别精度。 reinforcement learning deep reinforcement learning

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

#题目一句话要点标签🔗⭐
32 Handle Object Navigation as Weighted Traveling Repairman Problem 提出WTRP-Searcher，将零样本物体导航建模为加权旅行维修工问题 open-vocabulary open vocabulary foundation model
33 AKF-LIO: LiDAR-Inertial Odometry with Gaussian Map by Adaptive Kalman Filter 提出自适应卡尔曼滤波器以解决LiDAR-惯性测量系统的噪声估计问题 LIO ✅

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
34 Learning Physics-Based Full-Body Human Reaching and Grasping from Brief Walking References 提出一种基于少量行走数据学习物理可行的全身抓取运动生成框架 motion generation

⬅️ 返回 cs.RO 首页 · 🏠 返回主页