cs.RO（2026-09-23）

📊 共 48 篇论文 | 🔗 6 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (33 🔗5) 支柱二：RL算法与架构 (RL & Architecture) (6 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (5) 支柱三：空间感知与语义 (Perception & Semantics) (4)

🔬 支柱一：机器人控制 (Robot Control) (33 篇)

#题目一句话要点标签🔗⭐
1 TANDEM: Task and Motion Planning with As-Needed Demonstrations for Efficient Vision-Language-Action Model Fine-tuning 提出TANDEM以解决机器人演示数据收集效率低下问题 manipulation motion planning teleoperation
2 ForgetMimic: Motion Unlearning for Reinforcement Learning Humanoid Control 提出ForgetMimic以解决人形机器人控制中的运动遗忘问题 humanoid humanoid robot humanoid control
3 LiMA: Bridging Long-term Imagination to Real-time Dexterous Manipulation via Asynchronous Diffusion 提出LiMA以解决长时间想象与实时灵巧操控的协调问题 manipulation dexterous manipulation bi-manual ✅
4 InternW0: A Foundational Physical World Model for Efficient Real-World Interactions 提出InternW0以实现高效的现实世界交互 manipulation dexterous manipulation flow matching
5 BEE: Intervention-Adaptive Real-World Reinforcement Learning with Vision-Language-Action Models 提出BEE框架以解决现实世界强化学习中的人类干预问题 manipulation reinforcement learning vision-language-action
6 PointCast: One World Model for Rigid, Articulated, and Deformable Object Manipulation 提出PointCast以解决多种物体操控的世界模型问题 manipulation teleoperation world model
7 Distillation for Efficient Multitask Manipulation Policies via Conditional Flow Matching 通过条件流匹配蒸馏实现高效的多任务操作策略 manipulation policy learning flow matching
8 Dissecting Advantage-Guided Post-Training for Vision-Language-Action Policies 提出优势引导的后训练方法以优化视觉-语言-动作策略 bi-manual reinforcement learning vision-language-action
9 VLMs Can Describe, But Not Measure: Object-Centric Scene Understanding for Robotic Manipulation 提出VLM驱动的模块化感知框架以解决机器人操作中的场景理解问题 manipulation depth estimation scene understanding
10 Learning a Speed-adaptive Hip Exoskeleton Control Policy Via Sim-to-real Reinforcement Learning 提出基于仿真到现实的强化学习框架以优化个性化外骨骼控制策略 sim-to-real reinforcement learning preference learning
11 Talk2Escape: Conversational Grounding for Vision-and-Language Navigation 提出Talk2Escape以解决视觉语言导航中的错误恢复问题 quadruped sim-to-real Unitree
12 MemBodied: Recurrent Associative Memory for Vision-Language-Action Models 提出MemBodied以解决历史依赖任务中的记忆问题 manipulation vision-language-action
13 X2Real: an eXtensive simulation benchmark for real-world generalist policies 提出X2Real以解决机器人通用操作策略评估中的挑战 manipulation bi-manual sim-to-real
14 CoRe-WAM: Correspondence-Aligned Temporal Residuals for World Action Models 提出CoRe-WAM以解决机器人场景变化理解问题 manipulation world action model world action models
15 Banana Kick: Response-Informed Skill Evolution for Humanoid Soccer 提出响应信息驱动的技能演化方法以解决类人足球踢球问题 humanoid sim-to-real reinforcement learning ✅
16 Amplify: A Lightweight Library for Reproducible Nonlinear Programming Problems in Robotics 提出Amplify库以解决机器人非线性优化问题的可重复性 bipedal biped locomotion
17 Contact-Implicit Stein Projected ADMM for Discovery of Diverse Contact-Rich Manipulation Strategies 提出接触隐式Stein投影ADMM以发现多样化的接触丰富操作策略 manipulation trajectory optimization ✅
18 BrickCraft-Duo: Efficient Dual-Arm Skill Learning and Refinement for Compositional Long-Horizon Assembly 提出BrickCraft-Duo以解决双臂协作装配中的技能学习问题 manipulation dual-arm ✅
19 DAVIS: A Depth-Only End-to-End Active-Vision Framework for Humanoid Soccer Skills 提出DAVIS框架以解决人形机器人足球技能学习问题 humanoid AMP
20 InfiNoVA: Infinite Novel View Augmentation for Viewpoint Invariant Robot Policies 提出InfiNoVA以解决机器人策略对视角依赖的问题 manipulation vision-language-action VLA
21 Gray-Box Model Predictive Control for Articulated Dump Trucks via Gaussian Process Learning of Sideslip 提出高斯过程学习侧滑角的灰箱模型预测控制以提升自卸卡车精度 MPC model predictive control
22 A Modular Dual-Arm Robotic Cell for Disassembly and Repair of Industrial Control Electronics 提出模块化双臂机器人单元以解决工业控制电子设备的拆解与修复问题 dual-arm teleoperation
23 Safety-Filtered Distributed Koopman-MPC 提出安全过滤的分布式Koopman-MPC以解决通信丢包问题 MPC model predictive control
24 Omnidirectional Amphibious Locomotion via Internal Mass Actuation 提出MARBLE以解决多种地形和水面移动问题 locomotion
25 A Quasi-Direct-Drive Underactuated Asymmetric Hand for Dexterous and Efficient Grasping and Manipulation 提出一种准直接驱动的非完全驱动不对称手以解决灵巧抓取问题 manipulation ✅
26 LEAP-CBF: A Safety Filter for Uncertain Systems with Least-Effort Adversarial Potentials 提出LEAP-CBF以解决不确定系统的安全过滤问题 quadruped reinforcement learning deep reinforcement learning
27 EmbodiedSWE: Coding Agents for Long Horizon Dexterous Robotics 提出EmbodiedSWE以解决长时间复杂机器人任务的编码问题 manipulation VLA
28 Watch, Recall, Act: Always-On Robots in Concurrent Embodied Streams 提出ARMS以解决永续机器人在多模态流中的决策问题 dual-arm teleoperation
29 CoPRE: Improving Sensitivity in Proprioceptive Contact Detection for Low-Cost Robot Arms 提出CoPRE以提高低成本机器人臂的自我感知接触检测灵敏度 manipulation Unitree
30 Memory That Changes Action Is Not Memory That Guides It: Counterfactual Auditing of History-Conditioned Robot Policies 提出反事实记忆审计以解决机器人决策中的记忆指导问题 manipulation dual-arm
31 GLoTouch: Global-to-Local Haptic Perception Using a Parallel Gripper for Object Search, Recognition, and Grasping Without External Vision 提出GLoTouch以解决低光环境下的物体感知与抓取问题 manipulation
32 RegenHarness: A Robot Agent Harness with Evidence-Gated Recursive Self-Improvement 提出RegenHarness以解决长时间机器人执行中的任务完成问题 quadruped
33 BladeMaster: Real-Time Robotic Cutting Simulation with Online-Generated Persistent Discontinuities 提出BladeMaster以解决机器人切割仿真中的拓扑变化问题 manipulation

🔬 支柱二：RL算法与架构 (RL & Architecture) (6 篇)

#题目一句话要点标签🔗⭐
34 Tractable Reinforcement Learning for Full Class of Signal Temporal Logic Specifications Using Spatiotemporal Tube Reward 提出时变几何奖励的强化学习框架以解决复杂控制问题 reinforcement learning SAC spatiotemporal
35 Generalizable Robotic Insertion with World Models 提出基于世界模型的通用机器人插入框架以解决多样化部件组装问题 world model world models
36 DUGM-R: Uncertainty-Aware Dynamic Grid Mapping and Risk-Triggered Recovery for Learned Local Navigation 提出不确定性感知动态网格映射以解决室内导航中的碰撞风险问题 reinforcement learning motion estimation
37 Behavior-Aligned Action Tokenization for Robot Policy Learning 提出行为对齐动作标记化方法以提升机器人策略学习 policy learning
38 Turning Safety into Competence: Minimally Exploitable Robot Policies via Safety-Filtered Reinforcement Learning 提出S2C框架以解决机器人安全与竞争任务的矛盾 reinforcement learning
39 DEAL-Grasp: Decoupled Alignment Representation for Geometry-Aware Dexterous Grasp Generation 提出DEAL-Grasp以解决虚拟现实中的灵巧抓取问题 flow matching penetration ✅

🔬 支柱九：具身大模型 (Embodied Foundation Models) (5 篇)

#题目一句话要点标签🔗⭐
40 Less Language, More Latents: Annotation-Efficient VLAs for Driving 提出LADA以解决自动驾驶中的语言标注稀缺问题 vision-language-action VLA language conditioned
41 Multimodal Voice Activity Projection for Social Robot Mediation: Expected Behavior and Deployment Constraints 提出多模态语音活动预测以优化社交机器人调解行为 multimodal
42 Where Should I Join? Robot Group Joining via Language-Guided Goal Prediction 提出语言引导的机器人群体加入方法以解决社交导航问题 multimodal language conditioned
43 Controlling Collectives of AI Agents in Reasoning Space with Spatial Transformers 提出COMPASS以解决多机器人协作中的控制问题 large language model
44 RoboCafé in the Open: Interaction Continuity in Long-Term Public Human-Robot Interaction 提出RoboCafé以解决长期公共人机交互中的互动连续性问题 multimodal

🔬 支柱三：空间感知与语义 (Perception & Semantics) (4 篇)

#题目一句话要点标签🔗⭐
45 A Sample-Based Approach for Hierarchical Information-Theoretic Compression of Probabilistic Occupancy Grids 提出基于样本的方法以解决大规模概率占用网格的压缩问题 occupancy grid
46 NaviScale: Generating Large-Scale Semantic Map Datasets for Object Navigation 提出NaviScale以解决大规模语义地图数据集生成问题 semantic map
47 Spatial and Semantic Reasoning for LLM-Driven Robot Navigation via MCP 提出基于MCP的框架以解决LLM与ROS导航集成问题 occupancy grid large language model
48 DAVIO: Dense Monocular-Inertial SLAM with Feed-Forward Initialization and Pose-Conditioned Mapping 提出DAVIO以解决稀疏特征映射问题 VIO Depth Anything

⬅️ 返回 cs.RO 首页 · 🏠 返回主页