cs.RO（2025-09-25）

📊 共 34 篇论文

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (24) 支柱三：空间感知与语义 (Perception & Semantics) (4) 支柱二：RL算法与架构 (RL & Architecture) (3) 支柱九：具身大模型 (Embodied Foundation Models) (2) 支柱四：生成式动作 (Generative Motion) (1)

🔬 支柱一：机器人控制 (Robot Control) (24 篇)

#题目一句话要点标签🔗⭐
1 Joint Flow Trajectory Optimization For Feasible Robot Motion Generation from Video Demonstrations 提出Joint Flow Trajectory Optimization框架，解决视频示教中机器人运动规划问题 manipulation trajectory optimization teleoperation
2 AnywhereVLA: Language-Conditioned Exploration and Mobile Manipulation AnywhereVLA：面向复杂环境的语言条件移动操作框架 manipulation mobile manipulation reachability-aware
3 SEEC: Stable End-Effector Control with Model-Enhanced Residual Learning for Humanoid Loco-Manipulation 提出SEEC框架，通过模型增强残差学习实现人型机器人稳定末端执行器控制 humanoid humanoid robot bipedal
4 Cross-Modal Instructions for Robot Motion Generation 提出CrossInstruct框架，利用跨模态指令生成机器人运动轨迹 teleoperation reinforcement learning motion generation
5 RuN: Residual Policy for Natural Humanoid Locomotion 提出RuN残差策略，实现自然人型机器人步跑切换与高速运动控制 humanoid humanoid robot humanoid locomotion
6 RobotDancing: Residual-Action Reinforcement Learning Enables Robust Long-Horizon Humanoid Motion Tracking 提出基于残差动作强化学习的RobotDancing框架，实现鲁棒的人形机器人长时程运动跟踪。 humanoid sim-to-real Unitree
7 MPC-based Deep Reinforcement Learning Method for Space Robotic Control with Fuel Sloshing Mitigation 提出基于MPC的深度强化学习方法，用于抑制燃料晃动的空间机器人控制 MPC model predictive control reinforcement learning
8 Multi-Robot Vision-Based Task and Motion Planning for EV Battery Disassembly and Sorting 提出基于视觉的多机器人任务与运动规划框架，用于电动汽车电池拆卸与分拣。 motion planning task and motion planning TAMP
9 Towards Versatile Humanoid Table Tennis: Unified Reinforcement Learning with Prediction Augmentation 提出基于预测增强的统一强化学习框架，实现通用人形机器人乒乓球控制 humanoid locomotion reinforcement learning
10 BiNoMaP: Learning Category-Level Bimanual Non-Prehensile Manipulation Primitives 提出BiNoMaP，学习类别级双臂非抓取操作原语，提升机器人操作泛化性。 manipulation bi-manual dual-arm
11 RetoVLA: Reusing Register Tokens for Spatial Reasoning in Vision-Language-Action Models RetoVLA：通过复用Register Tokens增强VLA模型在机器人操作中的空间推理能力 manipulation vision-language-action VLA
12 Suction Leap-Hand: Suction Cups on a Multi-fingered Hand Enable Embodied Dexterity and In-Hand Teleoperation 提出基于吸盘的多指灵巧手SLeap Hand，实现超越人手的灵巧操作与遥操作 manipulation in-hand manipulation sim-to-real
13 Learning Terrain-Specialized Policies for Adaptive Locomotion in Challenging Environments 提出地形 специализирани политики的分层强化学习框架，提升复杂地形下的机器人运动能力 legged robot locomotion reinforcement learning
14 Autonomous UAV-Quadruped Docking in Complex Terrains via Active Posture Alignment and Constraint-Aware Control 提出主动姿态对齐和约束感知控制的无人机-四足机器人复杂地形自主对接框架 quadruped reinforcement learning deep reinforcement learning
15 ImaginationPolicy: Towards Generalizable, Precise and Reliable End-to-End Policy for Robotic Manipulation 提出基于运动导向关键点链的机器人操作端到端策略，提升泛化性、精度和可靠性。 manipulation VLA
16 DAGDiff: Guiding Dual-Arm Grasp Diffusion to Stable and Collision-Free Grasps 提出DAGDiff以解决双臂抓取的稳定性与碰撞问题 dual-arm
17 Efficient Differentiable Contact Model with Long-range Influence 提出高效可微接触模型，解决梯度消失问题，提升机器人控制性能 locomotion manipulation model predictive control
18 EEG-Driven AR-Robot System for Zero-Touch Grasping Manipulation 提出基于脑电的AR-机器人系统，实现零接触抓取操作 manipulation
19 Cyber Racing Coach: A Haptic Shared Control Framework for Teaching Advanced Driving Skills 提出一种触觉共享控制框架，用于提升驾驶员的高性能驾驶技能。 shared control
20 Wall Inspector: Quadrotor Control in Wall-proximity Through Model Compensation 提出基于吸力补偿模型预测控制的四旋翼近墙稳定控制方法 MPC model predictive control MAE
21 SLAM-Free Visual Navigation with Hierarchical Vision-Language Perception and Coarse-to-Fine Semantic Topological Planning 提出一种基于视觉-语言分层感知和粗细粒度语义拓扑规划的无SLAM视觉导航框架 legged robot locomotion reinforcement learning
22 Generating Stable Placements via Physics-guided Diffusion Models 提出基于物理引导扩散模型的稳定放置生成方法，提升机器人操作的鲁棒性和效率。 manipulation penetration
23 FSGlove: An Inertial-Based Hand Tracking System with Shape-Aware Calibration FSGlove：一种基于惯性传感器的手部追踪系统，通过形状感知校准实现高精度动作捕捉。 manipulation MANO
24 Developing a Mono-Actuated Compliant GeoGami Robot 提出一种单驱动柔性GeoGami机器人，实现形状变换和移动 locomotion

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

#题目一句话要点标签🔗⭐
25 Efficient Construction of Implicit Surface Models From a Single Image for Motion Generation FINS：基于单张图像快速构建隐式表面模型，用于机器人运动生成。 implicit representation motion generation foundation model
26 Digital Twin-Guided Robot Path Planning: A Beta-Bernoulli Fusion with Large Language Model as a Sensor 提出基于数字孪生的机器人路径规划方法，融合LLM语义信息提升安全性 semantic map large language model
27 Real-Time Indoor Object SLAM with LLM-Enhanced Priors 提出LLM增强先验的实时室内物体SLAM，解决稀疏观测下的优化难题 scene understanding large language model
28 SemSight: Probabilistic Bird's-Eye-View Prediction of Multi-Level Scene Semantics for Navigation SemSight：用于导航的多层次场景语义概率鸟瞰图预测 semantic map egocentric

🔬 支柱二：RL算法与架构 (RL & Architecture) (3 篇)

#题目一句话要点标签🔗⭐
29 Self-CriTeach: LLM Self-Teaching and Self-Critiquing for Improving Robotic Planning via Automated Domain Generation Self-CriTeach：利用LLM自教自评改进机器人规划，实现自动领域生成 reinforcement learning large language model chain-of-thought
30 Normalizing Flows are Capable Visuomotor Policy Learning Models 提出基于Normalizing Flows的视觉运动策略学习模型，提升推理效率与置信度评估。 policy learning diffusion policy
31 Leveraging Temporally Extended Behavior Sharing for Multi-task Reinforcement Learning 提出MT-Lévy，结合行为共享与时间扩展探索，提升多任务强化学习在机器人领域的样本效率。 reinforcement learning

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

#题目一句话要点标签🔗⭐
32 Automotive-ENV: Benchmarking Multimodal Agents in Vehicle Interface Systems 提出Automotive-ENV：用于车载界面多模态智能体的基准测试与交互环境 multimodal
33 Building Information Models to Robot-Ready Site Digital Twins (BIM2RDT): An Agentic AI Safety-First Framework BIM2RDT：构建信息模型到机器人就绪的施工现场数字孪生，保障安全 large language model

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
34 Taxonomy-aware Dynamic Motion Generation on Hyperbolic Manifolds 提出基于双曲流形的分类感知动态运动生成模型，用于生成类人机器人动作 motion generation human motion

⬅️ 返回 cs.RO 首页 · 🏠 返回主页