cs.RO（2026-09-24）

📊 共 73 篇论文 | 🔗 9 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (54 🔗7) 支柱二：RL算法与架构 (RL & Architecture) (11 🔗1) 支柱三：空间感知与语义 (Perception & Semantics) (5) 支柱九：具身大模型 (Embodied Foundation Models) (3 🔗1)

🔬 支柱一：机器人控制 (Robot Control) (54 篇)

#题目一句话要点标签🔗⭐
1 BeyondRetarget: Learning Executable Humanoid Motions Directly from Monocular Video 提出BeyondRetarget以解决人类视频到机器人动作映射问题 humanoid humanoid robot locomotion
2 DAWN: Noise-Robust Quadruped Parkour via Depth-Denoising World Models 提出DAWN框架以解决四足机器人在噪声环境下的运动问题 quadruped legged locomotion locomotion
3 TactileStep: Sole Tactile Learning for Regulating Foot-Terrain Interaction in Humanoid Locomotion 提出TactileStep以解决类人机器人足部与地面交互问题 humanoid humanoid robot humanoid locomotion
4 Res-HIL: Human-Guided Residual Reinforcement Learning for Sample-Efficient Dexterous Manipulation 提出Res-HIL框架以提高样本效率的灵巧操作学习 manipulation dexterous manipulation reinforcement learning
5 CrossSafe: Towards Cross-Embodiment Latent Safety Filters 提出CrossSafe以解决跨实体安全过滤问题 manipulation bi-manual cross-embodiment
6 Coding Agents for Generalized Task and Motion Planning Problems 提出编码代理以解决广义任务与运动规划问题 motion planning task and motion planning TAMP
7 Self-Adaptive VLA for Robust Robot Deployment 提出自适应VLA以解决机器人部署中的硬件适应问题 manipulation dexterous manipulation bi-manual ✅
8 Underwater C3-JEPA: An Object-Centric Cross-View World Model for ROV Salvage 提出Underwater C3-JEPA以解决ROV水下打捞中的对象状态预测问题 MPC model predictive control world model
9 AdaHVLA: Adaptive Harnesses for Long-Horizon Vision-Language-Action Execution 提出AdaHVLA以解决长时间视语言行动执行中的记忆与规划问题 manipulation vision-language-action VLA
10 Contact as a Decision Variable: Capability-Tradeoff Contact Selection for Legged Loco-Manipulation 提出能力权衡接触选择方法以优化腿部运动操控任务 quadruped manipulation loco-manipulation
11 Echo in the Steps: Learning Perceptive Humanoid Parkour with Gated Memory 提出感知人形机器人跑酷框架以解决复杂环境中的稳定行走问题 humanoid humanoid robot locomotion
12 CAMP: Cooperative Arm-Hand Motion Planning in Constrained Spaces 提出CAMP以解决复杂环境中的协调手臂-手部运动规划问题 manipulation motion planning motion generation ✅
13 Rolling-WAM: World Action Models with Rolling Imagination 提出Rolling-WAM以解决机器人操作中的延迟问题 humanoid manipulation Unitree
14 World Action Agent: Harnessing VLMs for Robot Manipulation via World Action Rehearsal 提出世界行动代理以解决机器人操控中的视觉语言模型应用问题 manipulation VLA multimodal
15 ActGaze: Learning Action-Grounded Gaze through Counterfactual Visual Interventions for High-Precision Manipulation 提出ActGaze以解决高精度机器人操作中的视觉注意力分散问题 manipulation vision-language-action VLA
16 Online Sim-to-Real Adaptation via Closed-Loop System Modeling 提出OSRAM框架以解决Sim-to-Real适应问题 bipedal biped manipulation
17 PolyUMI: Accessible Visual-Tactile-Audio Data Collection for Object Inference and Manipulation 提出PolyUMI与VisTA以增强机器人对物体操作的感知能力 manipulation imitation learning contact-aware
18 Robo-Harness K1: Harnessing Robot-Use Agents via Perception Augmentation 提出Robo-Harness K1以增强机器人使用代理的感知能力 manipulation dual-arm vision-language-action
19 Streaming-WAM: Action-Conditioned World-Action Model for Asynchronous Robot Manipulation 提出Streaming-WAM以解决异步机器人操作中的视觉预测问题 manipulation world model world models
20 Coupled State-Space Modelling, Control, and Policy Distillation for Hybrid Rigid-Pneumatic Manipulators 提出耦合状态空间建模与控制方法以优化混合刚性-气动操纵器 MPC model predictive control reinforcement learning
21 Faster Visuomotor Policy Learning on Action Manifolds via Riemannian MeanFlow 提出Riemannian MeanFlow策略以解决机器人动作生成效率问题 manipulation policy learning flow matching
22 Representation World Model: Learning States, Transition and Executable Plans in Representation 提出Representation World Model以解决传统世界模型规划问题 manipulation world model world models
23 Sim-to-Real Aware End-to-End Learning Environment for Micromobility 提出一种针对微型出行的端到端学习环境以解决仿真与现实间差距问题 sim-to-real world model world models
24 CALM: Current Aligned Link Manipulation for Single Arm Oversized Object Lifting 提出当前对齐链接操控以解决单臂大型物体搬运问题 manipulation sim-to-real actuator dynamics
25 EgoSpeedUp: Transferring Human Manipulation Tempo to Robot Policies 提出EgoSpeedUp以解决机器人操作节奏不足问题 manipulation imitation learning behavior cloning
26 Simple Torque-Observation Alignment for Zero-Shot Sim-to-Real Grasping with a Direct-Drive Gripper 提出简单的扭矩观测对齐方法以解决零-shot仿真到真实抓取问题 sim-to-real reinforcement learning teacher-student
27 ReVAMP: Vector-Accelerated Motion Planning for Kinematically-Constrained Systems via Reparameterization 提出ReVAMP以解决运动规划中的约束问题 manipulation motion planning
28 From Passive Execution to Active Exploration: Agentic Embodied Manipulation in Realistic Environments 提出主动探索框架以解决被动执行的机器人操作问题 manipulation scene understanding
29 UCON: Uncertainty-aware Navigation with Historical Re-association in Dynamic Environments 提出UCON以解决动态环境中的导航不确定性问题 trajectory optimization motion estimation
30 ADM-Planner: LLM-Guided Long-Horizon Planning for Mobile Manipulators with Attention-Enhanced Dynamic Memory 提出ADM-Planner以解决动态环境中的长时间规划问题 manipulation mobile manipulation large language model ✅
31 Self-Supervised Anchoring of Fingertip Sensing to Proprioception and Proactive Actions for Robot Imitation Learning 提出PROPRA方法以解决机器人模仿学习中的传感器信息利用问题 manipulation imitation learning ✅
32 A Tendon-Driven Robotic Jellyfish with Constrained Soft Actuation and Depth Control via Reinforcement Learning 提出一种基于腱驱动的机器人水母以解决水下运动控制问题 locomotion reinforcement learning
33 A Field-Deployable GNSS-based Navigation Stack for Outdoor Mobile Robots 提出基于GNSS的导航系统以解决户外移动机器人导航问题 model predictive control MAE geometric consistency ✅
34 Body-Grounded Replanning for Physically Adaptive Manipulation 提出基于身体状态的高层重规划以解决机器人操作适应性问题 manipulation
35 A Simple Gripper Interface for Simulator-Agnostic Cloth Manipulation 提出一种简单的抓取接口以解决布料操作问题 manipulation
36 Dense-Joint-Based Obstacle-Aided Locomotion with a Joint-Repositionable Snake Robot 提出高密度关节蛇形机器人以解决障碍物辅助运动问题 locomotion
37 Anthropomimetic Soft Robotic Forearm with Independently Articulated Carpal Bones Enabling Human-Like Adaptive Stiffness Modulability 提出类人软体机器人前臂以实现人类腕关节的适应性刚度调节 humanoid humanoid robot manipulation
38 RAPID: Robot Agentic Programming from Demonstrations 提出RAPID以解决机器人编程自动化问题 manipulation trajectory optimization
39 A Support-Enhanced Granular-Jamming Gripper for RL-based Grasping with Continuum Manipulators 提出支持增强颗粒夹持器以解决连续操纵器抓取问题 manipulation reinforcement learning
40 WRAP: Fixtureless Wrench-aware Multi-Robot Assembly Planning 提出WRAP以解决无夹具多机器人装配规划问题 motion planning
41 Koopman-Accelerated Model-Based Diffusion for Real-Time Robot Control 提出BK-MBD以解决传统模型基础扩散的计算瓶颈问题 trajectory optimization
42 HuGo: LLMs as Whole-Body Policy Code Designers for Humanoid Loco-Manipulation 提出HuGo以解决人形机器人运动与操作任务的复杂性问题 humanoid locomotion manipulation
43 Aerial Manipulation in the Wild with Onboard Perception, Policy Learning, and Whole-Body Control 提出一种新框架以解决户外环境中的空中操控问题 whole-body control manipulation MPC
44 Privacy-Preserving Prompted Policy Search for Robotic Control 提出隐私保护的提示策略搜索以解决机器人控制中的数据泄露问题 locomotion manipulation reinforcement learning
45 POIL: Point-based One-Shot Imitation Learning with Stable Dynamical Systems 提出POIL框架以解决一-shot模仿学习中的动态执行问题 manipulation imitation learning large language model ✅
46 EgoSpeedUp: Transferring Human Manipulation Tempo to Robot Policies 提出EgoSpeedUp以解决机器人操作节奏问题 manipulation imitation learning behavior cloning
47 DGT-Map: Directional Global Traversability Mapping Utilizing Multi-Task Learning for Heterogeneous Vehicles 提出DGT-Map以解决异构车辆的方向性越野可 traversability问题 locomotion traversability
48 GraspTwin: Zero-Shot Task-Oriented Grasp Optimization via a Digital Twin 提出GraspTwin以解决零-shot任务导向抓取优化问题 sim-to-real affordance foundation model ✅
49 Audit Before You Commit: Locating Belief Failures in Active Identification for One-Shot Manipulation 提出审计机制以解决机器人操作中的信念失败问题 manipulation
50 Learning-Accelerated Narrow-Phase Collision Detection via Check Ordering for Sampling-Based Motion Planning 提出基于学习的检查顺序优化以加速碰撞检测 motion planning
51 Tactile Sensing Array for Multi-Phalanx Sensing in Humanoid Hands 提出低成本触觉传感阵列以解决类人手抓取反馈不足问题 humanoid
52 Containing Behavioral Cascades from Manipulated Claims in LLM-Powered Multi-Robot Systems 提出主动验证框架以解决多机器人系统中的行为级联问题 manipulation large language model
53 A Support-Enhanced Granular-Jamming Gripper for RL-based Grasping with Continuum Manipulators 提出支持增强颗粒夹持器以解决连续操纵器抓取问题 manipulation reinforcement learning
54 Learning-Based Pressure Predictive Control of a Vertebraic Soft Robotic Tail 提出基于LSTM的压力预测控制以优化软体机器人尾巴运动 quadruped

🔬 支柱二：RL算法与架构 (RL & Architecture) (11 篇)

#题目一句话要点标签🔗⭐
55 Decoupled Early Exits for Task-Dependent Compute Allocation in Flow-Matching VLAs 提出解耦早期退出机制以优化流匹配VLA的计算分配 flow matching vision-language-action VLA
56 Continuous Online Fault Detection for Mobile Robots via Adaptive Edge Models 通过自适应边缘模型实现移动机器人连续在线故障检测 teacher-student distillation PULSE
57 M3GD: Multi-Modal Multi-View Geometric Diffusion for Camera--LiDAR Novel View Synthesis 提出M3GD以解决机器人新视图合成中的多模态融合问题 flow matching foundation model multimodal
58 From Target Selection to Digging: A Learning-Based Framework for Continuous Autonomous Excavation 提出基于学习的框架以解决连续自主挖掘问题 imitation learning elevation map multimodal
59 Training-free Behavior Cloning 提出行为预测控制方法以解决行为克隆训练成本高的问题 behavior cloning
60 Temperament Engineering: Designing Strategic Behavioural Diversity in Robot Swarms 提出气质工程以设计机器人群体的战略行为多样性 reinforcement learning foundation model
61 RACaP: Agentic Reasoning, Acting, and Coding as Policies for Evolvable Robot Learning 提出RACaP框架以解决机器人学习中的编码与决策问题 curriculum learning Code as Policies
62 HarnessPAI: An Evolving Harness for Physical AI 提出HarnessPAI框架以解决物理AI中的行为模型脆弱性问题 dreamer Code as Policies ✅
63 Outcome-Sensitive Motion Search for Impact-Aware Dexterous Catching 提出结果敏感运动搜索以解决影响感知抓取问题 reinforcement learning imitation learning
64 WALT: Learning World-Model-Aligned Latent Trajectories for Autonomous Driving 提出WALT以解决视觉世界模型与几何轨迹不匹配问题 world model world models JEPA
65 Policy-Calibrated DAgger: Offline Calibrated Noise Injection for Imitation Learning 提出Policy-Calibrated DAgger以解决模仿学习中的噪声注入问题 imitation learning diffusion policy multimodal

🔬 支柱三：空间感知与语义 (Perception & Semantics) (5 篇)

#题目一句话要点标签🔗⭐
66 FMCW-LIO: A Doppler LiDAR-Inertial Odometry 提出FMCW-LIO以解决传统LIO方法对几何特征的依赖问题 LIO
67 Markerless Multi-Modal Autonomous Robotic Inspection of Large Space Structures 提出无标记多模态自主机器人检查系统以解决大空间结构检测问题 3D reconstruction multimodal
68 OREN-X: Octree Residual Network for Real-Time Multi-Modal Mapping 提出OREN-X以解决多模态映射效率低下问题 open-vocabulary open vocabulary
69 FMCW-LIO: A Doppler LiDAR-Inertial Odometry 提出FMCW-LIO以解决传统LIO在动态环境中的局限性 LIO
70 VkVIO: Cross-platform GPU Acceleration for Visual-Inertial Odometry with Vulkan 提出VkVIO以解决视觉惯性里程计的跨平台加速问题 VIO

🔬 支柱九：具身大模型 (Embodied Foundation Models) (3 篇)

#题目一句话要点标签🔗⭐
71 GPT-6-Astra Lights Up Embodied Navigation: Evaluation in Zero-Shot Vision-and-Language Navigation in Continuous Environments 提出GPT-6-Astra以解决零-shot视觉语言导航问题 VLN foundation model
72 Actively Resolving Contextual Uncertainty for Underspecified Tasks in Natural Language 提出CLUE框架以解决自然语言下的不明确任务的上下文不确定性问题 foundation model language conditioned ✅
73 GPT-6-Astra Lights Up Embodied Navigation: Evaluation in Zero-Shot Vision-and-Language Navigation in Continuous Environments 提出GPT-6-Astra以解决零-shot视觉语言导航问题 VLN foundation model

⬅️ 返回 cs.RO 首页 · 🏠 返回主页