cs.RO（2025-09-15）

📊 共 21 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (17 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (2 🔗1) 支柱三：空间感知与语义 (Perception & Semantics) (1) 支柱七：动作重定向 (Motion Retargeting) (1)

🔬 支柱一：机器人控制 (Robot Control) (17 篇)

#题目一句话要点标签🔗⭐
1 TrajBooster: Boosting Humanoid Whole-Body Manipulation via Trajectory-Centric Learning TrajBooster：通过轨迹中心学习提升人形机器人全身操作能力 humanoid bipedal biped ✅
2 Tenma: Robust Cross-Embodiment Robot Manipulation with Diffusion Transformer Tenma：基于扩散Transformer的鲁棒跨具身机器人操作 manipulation bi-manual imitation learning
3 Embodied Navigation Foundation Model 提出跨具身、跨任务的导航基础模型NavFoM，提升具身智能导航的泛化能力。 quadruped cross-embodiment embodied AI
4 FR-Net: Learning Robust Quadrupedal Fall Recovery on Challenging Terrains through Mass-Contact Prediction FR-Net：基于质心-接触预测的四足机器人复杂地形跌倒恢复 quadruped legged robot fall recovery
5 Igniting VLMs toward the Embodied Space 提出WALL-OSS，提升VLM在具身环境中的理解与操作能力 manipulation foundation model multimodal
6 Learning Contact Dynamics for Control with Action-conditioned Face Interaction Graph Networks 提出基于动作条件面交互图网络的接触动力学学习模型，用于机器人操作控制。 manipulation MPC motion prediction
7 Inference-stage Adaptation-projection Strategy Adapts Diffusion Policy to Cross-manipulators Scenarios 提出一种推理阶段的自适应投影策略，使扩散策略适应跨机械臂场景 manipulation diffusion policy
8 Geometric Red-Teaming for Robotic Manipulation 提出几何红队（GRT）方法，通过几何扰动自动发现机器人操作策略的脆弱性。 manipulation
9 Augmented Reality-Enhanced Robot Teleoperation for Collecting User Demonstrations 提出基于增强现实的机器人遥操作系统，用于收集用户示教数据。 teleoperation
10 Time to Play: Simulating Early-Life Animal Dynamics Enhances Robotics Locomotion Discovery SMOL：通过模拟早期动物动态，提升机器人运动技能发现能力 locomotion
11 RAPTOR: A Foundation Policy for Quadrotor Control RAPTOR：一种用于四旋翼飞行器控制的基础策略，实现零样本泛化。 sim2real reinforcement learning imitation learning
12 Shape control of simulated multi-segment continuum robots via Koopman operators with per-segment projection 提出基于分段投影Koopman算子的软体机器人形状控制方法 MPC model predictive control
13 An integrated process for design and control of lunar robotics using AI and simulation 提出基于AI与仿真的月球机器人设计与控制集成流程 locomotion reinforcement learning
14 Design and Development of a Remotely Wire-Driven Walking Robot 提出远程钢丝驱动机制，实现恶劣环境下移动机器人的线缆遥控 quadruped
15 Bio-inspired tail oscillation enables robot fast crawling on deformable granular terrains 仿生尾部摆动提升机器人于可变形颗粒介质上的快速爬行能力 locomotion
16 Gesture-Based Robot Control Integrating Mm-wave Radar and Behavior Trees 提出基于毫米波雷达和行为树的手势控制机器人手臂系统，实现无接触人机交互。 manipulation
17 Adaptive Motorized LiDAR Scanning Control for Robust Localization with OpenStreetMap 提出基于OpenStreetMap引导的自适应激光雷达扫描控制，提升定位鲁棒性 model predictive control

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

#题目一句话要点标签🔗⭐
18 AssemMate: Graph-Based LLM for Robotic Assembly Assistance 提出AssemMate，利用图结构LLM辅助机器人进行装配任务。 large language model LLM for robotics ✅
19 PaiP: An Operational Aware Interactive Planner for Unknown Cabinet Environments 提出PaiP，解决未知橱柜环境中机器人操作的交互式规划问题 multimodal

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

#题目一句话要点标签🔗⭐
20 E2-BKI: Evidential Ellipsoidal Bayesian Kernel Inference for Uncertainty-aware Gaussian Semantic Mapping 提出E2-BKI，用于不确定性感知的室外环境高斯语义地图构建 semantic mapping semantic map sparse sensors

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

#题目一句话要点标签🔗⭐
21 UniPilot: Enabling GPS-Denied Autonomy Across Embodiments UniPilot：一种支持多种机器人平台在无GPS环境下自主导航的软硬件系统 cross-embodiment

⬅️ 返回 cs.RO 首页 · 🏠 返回主页