cs.RO（2025-10-16）

📊 共 30 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (22 🔗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) 支柱五：交互与反应 (Interaction & Reaction) (1) 支柱九：具身大模型 (Embodied Foundation Models) (1 🔗1) 支柱四：生成式动作 (Generative Motion) (1)

🔬 支柱一：机器人控制 (Robot Control) (22 篇)

#题目一句话要点标签🔗⭐
1 From Language to Locomotion: Retargeting-free Humanoid Control via Motion Latent Guidance RoboGhost：提出一种无重定向的语言引导人形机器人运动控制框架 humanoid humanoid robot humanoid control
2 VLA^2: Empowering Vision-Language-Action Models with an Agentic Framework for Unseen Concept Manipulation VLA^2：利用Agent框架增强VLA模型处理未见概念操作的能力 manipulation vision-language-action VLA
3 Open TeleDex: A Hardware-Agnostic Teleoperation System for Imitation Learning based Dexterous Manipulation Open TeleDex：一个硬件无关的灵巧操作模仿学习遥操作系统 manipulation dexterous hand dexterous manipulation
4 Towards Adaptable Humanoid Control via Adaptive Motion Tracking AdaMimic：基于自适应运动跟踪的通用人形机器人控制方法 humanoid humanoid robot humanoid control ✅
5 Architecture Is All You Need: Diversity-Enabled Sweet Spots for Robust Humanoid Locomotion 提出分层控制架构，提升人形机器人复杂地形的鲁棒运动性能 humanoid humanoid locomotion locomotion
6 Generative Models From and For Sampling-Based MPC: A Bootstrapped Approach For Adaptive Contact-Rich Manipulation 提出生成预测控制框架以提升接触丰富操作的采样效率 quadruped manipulation loco-manipulation
7 Risk-Aware Reinforcement Learning with Bandit-Based Adaptation for Quadrupedal Locomotion 提出基于Bandit自适应的风险感知强化学习，提升四足机器人运动鲁棒性 quadruped locomotion Unitree
8 Spatially anchored Tactile Awareness for Robust Dexterous Manipulation 提出SaTA框架，通过空间锚定的触觉感知实现鲁棒的灵巧操作 manipulation dexterous manipulation bi-manual
9 RM-RL: Role-Model Reinforcement Learning for Precise Robot Manipulation RM-RL：面向精准机器人操作的角色模型强化学习 manipulation reinforcement learning policy learning
10 SkyDreamer: Interpretable End-to-End Vision-Based Drone Racing with Model-Based Reinforcement Learning SkyDreamer：基于模型强化学习的可解释端到端视觉无人机竞速 sim-to-real reinforcement learning world model
11 Expertise need not monopolize: Action-Specialized Mixture of Experts for Vision-Language-Action Learning 提出AdaMoE，一种动作专用混合专家模型，提升VLA模型在机器人操作任务中的性能和效率。 manipulation vision-language-action VLA
12 RL-100: Performant Robotic Manipulation with Real-World Reinforcement Learning RL-100：基于真实世界强化学习的高性能机器人操作框架 manipulation reinforcement learning PPO
13 Leveraging Neural Descriptor Fields for Learning Contact-Aware Dynamic Recovery 提出CADRE框架，利用神经描述场学习接触感知的动态恢复，提升灵巧操作的鲁棒性。 manipulation dexterous manipulation reinforcement learning
14 VT-Refine: Learning Bimanual Assembly with Visuo-Tactile Feedback via Simulation Fine-Tuning VT-Refine：通过模拟微调学习基于视觉-触觉反馈的双臂装配 bi-manual sim-to-real reinforcement learning ✅
15 CBF-RL: Safety Filtering Reinforcement Learning in Training with Control Barrier Functions 提出CBF-RL框架，通过控制屏障函数在训练中安全过滤强化学习策略 humanoid humanoid robot Unitree
16 Restoring Noisy Demonstration for Imitation Learning With Diffusion Models 提出基于扩散模型的模仿学习框架，用于恢复含噪声的专家演示数据。 locomotion manipulation dexterous manipulation
17 Proprioceptive Image: An Image Representation of Proprioceptive Data from Quadruped Robots for Contact Estimation Learning 提出一种基于本体感受图像的四足机器人接触估计学习方法 quadruped locomotion
18 Learning Human-Humanoid Coordination for Collaborative Object Carrying 提出COLA算法，实现基于本体感觉的人形机器人协同搬运，提升人机协作效率。 humanoid reinforcement learning
19 Accelerated Multi-Modal Motion Planning Using Context-Conditioned Diffusion Models 提出CAMPD，利用上下文条件扩散模型加速多模态运动规划，提升泛化性。 motion planning
20 Prescribed Performance Control of Deformable Object Manipulation in Spatial Latent Space 提出一种基于空间潜在空间的柔性物体操作预定性能控制方法 manipulation
21 RDD: Retrieval-Based Demonstration Decomposer for Planner Alignment in Long-Horizon Tasks 提出RDD：一种基于检索的分解器，用于长时任务中规划器对齐 manipulation vision-language-action
22 When Planners Meet Reality: How Learned, Reactive Traffic Agents Shift nuPlan Benchmarks 引入SMART智能体，提升nuPlan自动驾驶规划器评估的真实性和可靠性 sim-to-real ✅

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

#题目一句话要点标签🔗⭐
23 Logic-based Task Representation and Reward Shaping in Multiagent Reinforcement Learning 提出基于逻辑的任务表示和奖励塑造方法，加速多智能体强化学习。 reinforcement learning reward shaping
24 EdgeNavMamba: Mamba Optimized Object Detection for Energy Efficient Edge Devices EdgeNavMamba：面向边缘设备的节能Mamba优化目标检测 reinforcement learning Mamba
25 SADCHER: Scheduling using Attention-based Dynamic Coalitions of Heterogeneous Robots in Real-Time SADCHER：基于注意力机制的异构多机器人实时动态联盟调度 imitation learning spatiotemporal

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

#题目一句话要点标签🔗⭐
26 QuASH: Using Natural-Language Heuristics to Query Visual-Language Robotic Maps QuASH：利用自然语言启发式方法查询视觉-语言机器人地图 scene understanding open-vocabulary open vocabulary
27 GOPLA: Generalizable Object Placement Learning via Synthetic Augmentation of Human Arrangement GOPLA：通过合成增强人类布置数据，学习可泛化的物体放置 affordance large language model

🔬 支柱五：交互与反应 (Interaction & Reaction) (1 篇)

#题目一句话要点标签🔗⭐
28 Stability Criteria and Motor Performance in Delayed Haptic Dyadic Interactions Mediated by Robots 针对时延触觉人机交互，提出机器人调解下的稳定性判据 dyadic interaction

🔬 支柱九：具身大模型 (Embodied Foundation Models) (1 篇)

#题目一句话要点标签🔗⭐
29 SUM-AgriVLN: Spatial Understanding Memory for Agricultural Vision-and-Language Navigation 提出SUM-AgriVLN，利用空间记忆提升农业视觉语言导航性能 VLN ✅

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
30 A Generalized Placeability Metric for Model-Free Unified Pick-and-Place Reasoning 提出一种广义可放置性指标，用于无模型统一抓取放置推理 physically plausible

⬅️ 返回 cs.RO 首页 · 🏠 返回主页