cs.RO（2026-09-04）

📊 共 17 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (9 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (3) 支柱三：空间感知与语义 (Perception & Semantics) (2) 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) (9 篇)

#题目一句话要点标签🔗⭐
1 Development of a Humanoid Robot Prototype for Multimodal Human-Robot Interaction 提出一种类人机器人原型以促进多模态人机交互 humanoid humanoid robot manipulation
2 Towards Neuro-Symbolic Procedural Reasoning for Long-Horizon Vision-Language-Action Manipulation 提出神经符号程序推理框架以解决长时间视觉-语言-动作操作问题 manipulation teleoperation policy learning
3 RoboSPA: Can VLA Models Go Beyond Simple Scenes and Short-Horizon Tasks? 提出RoboSPA以解决VLA模型在复杂场景中的推理问题 manipulation vision-language-action VLA ✅
4 What Matters, When? Diagnosing and Improving Conditional Visual Grounding in Visuomotor Imitation Policies 提出条件视觉定位方法以改善视觉运动模仿策略的性能 manipulation policy learning vision-language-action
5 Reasoning Without Inference Cost: Latent Semantic Scaffolding for Robot VLA Policies 提出潜在语义支架以降低机器人VLA策略推理成本 manipulation vision-language-action VLA
6 Temporal Tactile Encoding and Compliance for Intent-Aware Robot-to-Human Bimanual Handover 提出多模态手段以提升人机交互中的物品交接可靠性 bi-manual VLA multimodal
7 LIBERO-RECOVER: Beyond Task Success Towards Failure Recovery in Robotic Manipulation Models 提出LIBERO-Recover以解决机器人操作中的失败恢复问题 manipulation vision-language-action VLA ✅
8 Morphology and actuation as inductive biases in robotic hand manipulation 提出结构与驱动作为机器人手部操作的归纳偏置 manipulation dexterous hand reinforcement learning
9 TacPAC: Tactile Prediction and Real-Time Action Correction in World-Action Models for Contact-Rich Manipulation 提出TacPAC以解决接触丰富操作中的实时动作修正问题 manipulation ✅

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

#题目一句话要点标签🔗⭐
10 Dressing in Motion: A Human Motion-Aware Diffusion Policy for Robot-Assisted Dressing 提出人机运动感知扩散策略以解决机器人辅助穿衣问题 diffusion policy human motion
11 Coupled Control and Wireless World Models for Resilient Remote Robotic Control 提出基于耦合控制和无线世界模型的远程机器人控制框架以应对通信挑战 world model world models JEPA
12 Human-Human & Human-Robot Interaction Transformer (H2INT) for Robot Navigation in Dense and Uncertain Crowds 提出H2INT以解决密集人群中机器人导航安全问题 reinforcement learning policy learning interaction transformer

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

#题目一句话要点标签🔗⭐
13 NavArena: Automated Construction of Goal-Oriented Navigation Benchmarks from 3D Gaussian Splatting Reconstructions 提出NavArena以解决3D导航基准构建问题 3D gaussian splatting 3DGS gaussian splatting
14 Open-Set 3D Scene Graphs for Field Robotics: An Outdoor Case Study 提出开放集3D场景图以解决户外机器人导航问题 traversability

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

#题目一句话要点标签🔗⭐
15 One Word, Different Action: A Real-Robot Benchmark for Language-Conditioned Embodied Reasoning 提出基于语言指令的真实机器人基准以解决决策一致性与敏感性问题 language conditioned
16 A Schema Bounded Language Model for Refining Robot Policies Without Destabilizing Local Learning 提出一种模式约束语言模型以优化机器人策略而不干扰局部学习 large language model

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
17 FIRE-LIVWO: Robust LiDAR-Inertial-Visual-Wheel Odometry via Failure-Immune mmWave Radar Enhancement 提出FIRE-LIVWO以解决复杂地下煤矿中的鲁棒SLAM问题 penetration

⬅️ 返回 cs.RO 首页 · 🏠 返回主页