cs.RO（2026-06-08）

📊 共 33 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 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (5) 支柱九：具身大模型 (Embodied Foundation Models) (3) 支柱三：空间感知与语义 (Perception & Semantics) (2)

🔬 支柱一：机器人控制 (Robot Control) (23 篇)

#题目一句话要点标签🔗⭐
1 MotionWAM: Towards Foundation World Action Models for Real-Time Humanoid Loco-Manipulation 提出MotionWAM以解决实时人形机器人运动控制问题 humanoid humanoid control locomotion
2 TORL-VLA: Tactile Guided Online Reinforcement Learning for Contact-Rich Manipulation 提出TORL-VLA以解决接触丰富任务中的在线适应问题 manipulation reinforcement learning vision-language-action
3 ProbeAct: Probe-Guided Training-Free Failure Recovery in Vision-Language-Action Models 提出PROBEACT以解决VLA模型在操作失败时的恢复问题 manipulation vision-language-action VLA
4 ReCoVLA: VLM-Guided Reward Compilation for Failure Recovery in Vision-Language-Action Policies 提出ReCoVLA以解决视觉-语言-动作策略中的失败恢复问题 manipulation sim-to-real vision-language-action
5 MemoryVLA++: Temporal Modeling via Memory and Imagination in Vision-Language-Action Models 提出MemoryVLA++以解决机器人操控中的时间建模问题 manipulation world model world models ✅
6 $ω$-EVA: Envision, Verify, and Act with Latent Interactive World Models 提出$ω$-EVA以解决现有动作生成模型的局限性 bi-manual world model world models
7 Your Model Already Knows: Attention-Guided Safety Filter for Vision-Language-Action Models 提出基于注意力引导的安全过滤器以解决VLA模型的碰撞问题 manipulation vision-language-action VLA
8 Back to the Familiar Future: Failure Recovery for VLA Policies via Pre-Imagined Milestone Selection 提出B2FF框架以解决VLA策略的失败恢复问题 manipulation vision-language-action VLA
9 PTDL:Multi-Terrain Fall Recovery via Phase-Terrain Decoupled Learning 提出PTDL以解决多地形下人形机器人跌倒恢复问题 humanoid humanoid robot locomotion
10 iMaC: Translating Actions into Motion and Contact Images for Embodied World Models 提出iMaC以解决传统动作表示的局限性问题 manipulation world model world models
11 Safe Polytope-in-Polytope Motion Planning and Control with Control Barrier Functions 提出安全多面体运动规划方法以解决狭窄环境中的机器人控制问题 model predictive control motion planning occupancy grid
12 VAIC: Vision-Guided Humanoid Agile Object Interaction Control via Decoupled Commands 提出VAIC以解决人形机器人在复杂环境中的物体交互控制问题 humanoid humanoid robot distillation
13 SynManDex: Synthesizing Human-like Dexterous Grasps from Synthetic Human Pre-Grasps 提出SynManDex以解决机器人抓取中的人类手势转化问题 manipulation bi-manual affordance
14 Difference-Aware Retrieval Policies for Imitation Learning 提出差异感知检索策略以解决模仿学习中的泛化问题 manipulation imitation learning behavior cloning ✅
15 CT-VAM: A Cerebello-Thalamic-Inspired Vision-Action Model for Efficient Visuomotor Control 提出CT-VAM以解决高频低级执行中的任务意图处理问题 manipulation vision-language-action VLA
16 Autonomous Obstacle Removal for Excavators through Policy Learning with Particle Simulation 提出基于粒子模拟的自主障碍物移除策略以解决挖掘机自动化问题 sim-to-real policy learning curriculum learning
17 DexPIE: Stable Dexterous Policy Improvement from Real-World Experience 提出DexPIE以解决高维动作空间下的灵巧操控问题 manipulation dexterous hand dexterous manipulation
18 ReGIL: Retrieval-Guided Imitation Learning from a Single Demonstration 提出ReGIL框架以解决单一示范下机器人学习挑战 manipulation imitation learning
19 Motion planning for hundreds of floating robots 提出可扩展的运动规划方法以解决浮动机器人群体的碰撞避免问题 motion planning
20 Trajectory Optimization in Single and Dual-UAV Bearing-Only Target Localization 提出基于FIM的轨迹优化方法以解决无人机目标定位问题 trajectory optimization
21 Dense Force Estimation with an Event-based Optical Tactile Sensor 提出基于事件的光学触觉传感器的密集力估计方法 manipulation dexterous manipulation
22 AHA-WAM:Asynchronous Horizon-Adaptive World-Action Modeling with Observation-Guided Context Routing 提出AHA-WAM以解决世界-动作模型的时间耦合问题 manipulation policy learning
23 AetheRock: An Arm-Worn Robot Teaching System for Force-Guided Vision-Tactile Learning 提出AetheRock以解决力导向视觉触觉学习中的传感器不兼容问题 manipulation representation learning

🔬 支柱二：RL算法与架构 (RL & Architecture) (5 篇)

#题目一句话要点标签🔗⭐
24 Targeting World Models to Compromise Robot Learning Pipelines 提出针对世界模型的数据中毒攻击以解决机器人学习安全问题 DRL world model world models
25 Self-Paced Curriculum Reinforcement Learning for Autonomous Superbike Racing in Simulation 提出自适应课程强化学习以解决自主摩托车赛车问题 reinforcement learning deep reinforcement learning SAC
26 KPGrasp: Scalable Keypoint Flow Matching for Dexterous Grasp Generation 提出KPGrasp以解决高质量灵巧抓取生成问题 flow matching penetration
27 Shape Formation for the Cooperative Transportation of Arbitrary Objects Using Multi-Agent Reinforcement Learning 提出多智能体强化学习方法以解决复杂物体的协作运输问题 reinforcement learning
28 Physics-Aware Sparse Learning and Selective Online Adaptation for Euler-Lagrange Robot Dynamics 提出物理感知稀疏学习与选择性在线适应以解决机器人动力学建模问题 world model world models

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

#题目一句话要点标签🔗⭐
29 Harness Engineering for Physical AI: Robot Middleware Is the Harness Layer 提出机器人中间件作为物理人工智能的支撑层 vision-language-action VLA
30 From USD Scenes to Knowledge Graphs: Zero-Shot Ontology Grounding with LLMs 提出基于大语言模型的零-shot本体映射方法以解决知识图谱构建问题 large language model
31 SpaceVLN: A Zero-Shot Vision-and-Language Navigation Agent with Online Spatial Cognitive Memory and Reasoning 提出SpaceVLN以解决零-shot视觉语言导航问题 foundation model

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

#题目一句话要点标签🔗⭐
32 Autonomous FPV Flight with Translational Optical Flow and Uncertainty Mask 提出基于平移光流和不确定性掩码的自主FPV飞行方法以解决复杂环境中的飞行挑战 optical flow differentiable simulation
33 Dual Quaternion-Based Unscented Kalman Filter with Visual Inertial Odometry for Navigation in GPS-Denied Environments 提出双四元数无味卡尔曼滤波器以解决GPS缺失环境下导航问题 VIO

⬅️ 返回 cs.RO 首页 · 🏠 返回主页