cs.RO（2026-04-09）

📊 共 20 篇论文 | 🔗 1 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (14) 支柱九：具身大模型 (Embodied Foundation Models) (4) 支柱二：RL算法与架构 (RL & Architecture) (2 🔗1)

🔬 支柱一：机器人控制 (Robot Control) (14 篇)

#题目一句话要点标签🔗⭐
1 HEX: Humanoid-Aligned Experts for Cross-Embodiment Whole-Body Manipulation HEX：面向跨具身全身操作的人形对齐专家系统 humanoid humanoid robot humanoid control
2 Sumo: Dynamic and Generalizable Whole-Body Loco-Manipulation Sumo：基于预训练策略和测试时规划的通用动态全身操作 quadruped legged robot humanoid
3 EMMa: End-Effector Stability-Oriented Mobile Manipulation for Tracked Rescue Robots EMMa：面向末端执行器稳定性的履带式救援机器人移动操作 manipulation mobile manipulation motion generation
4 On-Policy Distillation of Language Models for Autonomous Vehicle Motion Planning 提出基于On-Policy蒸馏的语言模型，用于自动驾驶车辆运动规划，有效减小模型尺寸。 motion planning reinforcement learning distillation
5 ViVa: A Video-Generative Value Model for Robot Reinforcement Learning ViVa：一种用于机器人强化学习的视频生成价值模型 manipulation reinforcement learning spatiotemporal
6 Reset-Free Reinforcement Learning for Real-World Agile Driving: An Empirical Study 针对真实世界敏捷驾驶，提出无需重置的强化学习方法 sim-to-real reinforcement learning PPO
7 ActiveGlasses: Learning Manipulation with Active Vision from Ego-centric Human Demonstration ActiveGlasses：利用第一视角人类演示学习主动视觉操作 manipulation zero-shot transfer
8 Open-Ended Instruction Realization with LLM-Enabled Multi-Planner Scheduling in Autonomous Vehicles 提出基于LLM的多规划器调度框架，实现自动驾驶车辆中开放指令的执行。 MPC model predictive control large language model
9 SIM1: Physics-Aligned Simulator as Zero-Shot Data Scaler in Deformable Worlds SIM1：物理对齐的模拟器作为可变形物体世界中的零样本数据扩展器 manipulation sim-to-real policy learning
10 A-SLIP: Acoustic Sensing for Continuous In-hand Slip Estimation 提出A-SLIP：利用声学传感实现连续的机械手内滑移估计 manipulation in-hand manipulation
11 Density-Driven Optimal Control: Convergence Guarantees for Stochastic LTI Multi-Agent Systems 提出随机密度驱动最优控制以解决多智能体系统覆盖问题 MPC
12 Toward Hardware-Agnostic Quadrupedal World Models via Morphology Conditioning 提出基于形态条件化的四足机器人世界模型，实现硬件无关的通用运动控制。 quadruped legged locomotion locomotion
13 LEGO: Latent-space Exploration for Geometry-aware Optimization of Humanoid Kinematic Design LEGO：基于潜在空间探索的人形机器人运动学设计几何感知优化 humanoid motion retargeting human motion
14 SIM1: Physics-Aligned Simulator as Zero-Shot Data Scaler in Deformable Worlds SIM1：物理对齐的模拟器作为可变形世界中的零样本数据扩展器 manipulation sim-to-real policy learning

🔬 支柱九：具身大模型 (Embodied Foundation Models) (4 篇)

#题目一句话要点标签🔗⭐
15 Vision-Language Navigation for Aerial Robots: Towards the Era of Large Language Models 面向大语言模型的空中机器人视觉-语言导航综述 VLN large language model
16 Exploring Temporal Representation in Neural Processes for Multimodal Action Prediction 提出DMBN-PTE模型，提升机器人自监督多模态动作预测中时间表征的鲁棒性 multimodal
17 A Soft Robotic Interface for Chick-Robot Affective Interactions 提出一种面向雏鸡的软体机器人情感交互界面，提升动物-机器人交互的接受度。 multimodal
18 AgiPIX: Bridging Simulation and Reality in Indoor Aerial Inspection AgiPIX：用于室内飞行检测的模拟-现实桥接开放平台 zero-shot transfer

🔬 支柱二：RL算法与架构 (RL & Architecture) (2 篇)

#题目一句话要点标签🔗⭐
19 Incremental Residual Reinforcement Learning Toward Real-World Learning for Social Navigation 提出增量残差强化学习(IRRL)用于社交导航的真实世界机器人学习。 reinforcement learning deep reinforcement learning
20 RoboAgent: Chaining Basic Capabilities for Embodied Task Planning 提出RoboAgent，通过链式调用子能力解决具身任务规划中的长程推理问题。 reinforcement learning behavior cloning multimodal ✅

⬅️ 返回 cs.RO 首页 · 🏠 返回主页