cs.RO（2025-09-17）

📊 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) (23 🔗1) 支柱三：空间感知与语义 (Perception & Semantics) (4) 支柱二：RL算法与架构 (RL & Architecture) (4 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (2) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱一：机器人控制 (Robot Control) (23 篇)

#题目一句话要点标签🔗⭐
1 Behavior Foundation Model for Humanoid Robots 提出人形机器人行为基础模型，提升通用性和泛化能力 humanoid humanoid robot humanoid control
2 LeVR: A Modular VR Teleoperation Framework for Imitation Learning in Dexterous Manipulation LeVR：用于灵巧操作模仿学习的模块化VR遥操作框架 manipulation dexterous hand dexterous manipulation
3 Dynamic Adaptive Legged Locomotion Policy via Decoupling Reaction Force Control and Gait Control 提出解耦反应力控制与步态控制的动态自适应腿足运动策略，提升泛化性 legged locomotion locomotion locomotion policy
4 CRAFT: Coaching Reinforcement Learning Autonomously using Foundation Models for Multi-Robot Coordination Tasks CRAFT：利用具身智能自主指导多机器人强化学习，解决复杂协作任务 quadruped manipulation bi-manual
5 DreamControl: Human-Inspired Whole-Body Humanoid Control for Scene Interaction via Guided Diffusion DreamControl：通过引导扩散实现受人类启发的全身人形机器人场景交互控制 humanoid humanoid control sim-to-real ✅
6 SeqVLA: Sequential Task Execution for Long-Horizon Manipulation with Completion-Aware Vision-Language-Action Model SeqVLA：面向长时程操作任务，基于完成感知的视觉-语言-动作模型实现序列化任务执行 manipulation vision-language-action VLA
7 Whole-body Motion Control of an Omnidirectional Wheel-Legged Mobile Manipulator via Contact-Aware Dynamic Optimization 提出一种接触感知的全身动力学优化方法，用于全向轮腿式移动机械臂的运动控制。 quadruped legged robot locomotion
8 Dual-Actor Fine-Tuning of VLA Models: A Talk-and-Tweak Human-in-the-Loop Approach 提出基于强化学习的人机协同双Actor微调框架，提升VLA模型在复杂机器人任务中的性能。 manipulation reinforcement learning policy learning
9 CLAW: A Vision-Language-Action Framework for Weight-Aware Robotic Grasping CLAW：一种用于重量感知机器人抓取的视觉-语言-动作框架 manipulation dual-arm vision-language-action
10 Pre-Manipulation Alignment Prediction with Parallel Deep State-Space and Transformer Models 提出基于深度状态空间模型与Transformer的并行架构，预测操作任务中的预操作对齐情况。 manipulation open-vocabulary open vocabulary
11 FSR-VLN: Fast and Slow Reasoning for Vision-Language Navigation with Hierarchical Multi-modal Scene Graph 提出FSR-VLN，结合分层多模态场景图与快慢推理，提升视觉语言导航性能。 humanoid humanoid robot Unitree
12 PhysicalAgent: Towards General Cognitive Robotics with Foundation World Models PhysicalAgent：基于世界模型的通用认知机器人框架，实现迭代推理和闭环执行。 humanoid manipulation bi-manual
13 Multi-Quadruped Cooperative Object Transport: Learning Decentralized Pinch-Lift-Move 提出一种分散式策略，实现多足机器人协同搬运无抓取物体 quadruped locomotion sim2real
14 Language Conditioning Improves Accuracy of Aircraft Goal Prediction in Untowered Airspace 提出语言条件约束的飞机目标预测框架，提升非塔台空域自主飞行安全性 motion planning large language model multimodal
15 Reinforcement Learning for Robotic Insertion of Flexible Cables in Industrial Settings 提出基于强化学习和基础模型的柔性电缆机器人插入方法，实现零样本部署。 sim-to-real reinforcement learning foundation model
16 FlightDiffusion: Revolutionising Autonomous Drone Training with Diffusion Models Generating FPV Video FlightDiffusion：利用扩散模型生成FPV视频，革新无人机自主训练 sim-to-real policy learning physically plausible
17 SHaRe-RL: Structured, Interactive Reinforcement Learning for Contact-Rich Industrial Assembly Tasks SHaRe-RL：用于高精度工业装配的结构化交互式强化学习框架 manipulation reinforcement learning
18 Flexible and Foldable: Workspace Analysis and Object Manipulation Using a Soft, Interconnected, Origami-Inspired Actuator Array 提出一种基于软性互连折纸结构的柔性分布式操作器阵列，用于提升物体操作能力。 manipulation
19 Repulsive Trajectory Modification and Conflict Resolution for Efficient Multi-Manipulator Motion Planning 提出基于斥力轨迹修正的多机械臂运动规划方法，提升效率并解决冲突 motion planning
20 Track Any Motions under Any Disturbances Any2Track：强化学习框架，实现人形机器人复杂运动和抗扰动追踪 humanoid sim2real Unitree
21 RLBind: Adversarial-Invariant Cross-Modal Alignment for Unified Robust Embeddings RLBind：对抗不变跨模态对齐，用于统一鲁棒嵌入 manipulation zero-shot transfer
22 HGACNet: Hierarchical Graph Attention Network for Cross-Modal Point Cloud Completion HGACNet：用于跨模态点云补全的分层图注意力网络 manipulation spatial relationship
23 CDFlow: Generative Gradient Flows for Configuration Space Distance Fields via Neural ODEs CDFlow：利用神经ODE生成配置空间距离场的梯度流，提升高自由度机器人运动规划。 motion planning

🔬 支柱三：空间感知与语义 (Perception & Semantics) (4 篇)

#题目一句话要点标签🔗⭐
24 Perception-Integrated Safety Critical Control via Analytic Collision Cone Barrier Functions on 3D Gaussian Splatting 提出基于分析碰撞锥的安全控制方法以解决机器人避障问题 3D gaussian splatting 3DGS gaussian splatting
25 MCGS-SLAM: A Multi-Camera SLAM Framework Using Gaussian Splatting for High-Fidelity Mapping MCGS-SLAM：基于高斯溅射的多相机SLAM框架，实现高保真地图构建 3D gaussian splatting 3DGS gaussian splatting
26 BIM Informed Visual SLAM for Construction Monitoring 提出一种BIM约束的RGB-D SLAM系统，用于提升建筑工地监测精度。 visual SLAM
27 Barometer-Aided Attitude Estimation 提出气压计辅助的姿态估计算法，解决GNSS拒止环境下的倾角估计问题 visual odometry geometric consistency

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

#题目一句话要点标签🔗⭐
28 SEG-Parking: Towards Safe, Efficient, and Generalizable Autonomous Parking via End-to-End Offline Reinforcement Learning 提出SEG-Parking，通过端到端离线强化学习实现安全、高效、泛化的自动泊车 reinforcement learning offline RL offline reinforcement learning
29 Motion Adaptation Across Users and Tasks for Exoskeletons via Meta-Learning 提出基于元学习的外骨骼运动自适应方法，提升用户和任务泛化性 imitation learning motion adaptation
30 MIMIC-D: Multi-modal Imitation for MultI-agent Coordination with Decentralized Diffusion Policies 提出MIMIC-D以解决多模态多智能体协调问题 imitation learning diffusion policy
31 Aegis: Automated Error Generation and Attribution for Multi-Agent Systems Aegis：自动化生成并归因多智能体系统错误，提升系统鲁棒性。 reinforcement learning contrastive learning large language model ✅

🔬 支柱九：具身大模型 (Embodied Foundation Models) (2 篇)

#题目一句话要点标签🔗⭐
32 GeoAware-VLA: Implicit Geometry Aware Vision-Language-Action Model GeoAware-VLA：利用几何先验提升VLA模型视角泛化能力 vision-language-action VLA
33 GestOS: Advanced Hand Gesture Interpretation via Large Language Models to control Any Type of Robot GestOS：利用大语言模型实现手势对异构机器人团队的高级控制 large language model

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

#题目一句话要点标签🔗⭐
34 StableTracker: Learning to Stably Track Target via Differentiable Simulation 提出StableTracker，通过可微仿真学习稳定跟踪快速移动目标 differentiable simulation

⬅️ 返回 cs.RO 首页 · 🏠 返回主页