cs.RO（2026-04-07）

📊 共 70 篇论文 | 🔗 1 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (48 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (10) 支柱二：RL算法与架构 (RL & Architecture) (8) 支柱三：空间感知与语义 (Perception & Semantics) (1) 支柱八：物理动画 (Physics-based Animation) (1) 支柱七：动作重定向 (Motion Retargeting) (1) 支柱四：生成式动作 (Generative Motion) (1)

🔬 支柱一：机器人控制 (Robot Control) (48 篇)

#题目一句话要点标签🔗⭐
1 Dynamic Whole-Body Dancing with Humanoid Robots -- A Model-Based Control Approach 提出基于模型的控制框架，实现人形机器人的动态全身舞蹈动作生成与执行 humanoid humanoid robot MPC
2 RK-MPC: Residual Koopman Model Predictive Control for Quadruped Locomotion in Offroad Environments 提出RK-MPC：基于残差Koopman模型的四足机器人越野环境模型预测控制 quadruped locomotion MPC
3 ROSClaw: A Hierarchical Semantic-Physical Framework for Heterogeneous Multi-Agent Collaboration 提出ROSClaw框架，用于异构多智能体协作，弥合语义理解与物理执行的鸿沟。 manipulation sim-to-real policy learning
4 PALM: Progress-Aware Policy Learning via Affordance Reasoning for Long-Horizon Robotic Manipulation PALM：基于可供性推理和进度感知策略学习的长程机器人操作 manipulation policy learning affordance
5 ST-BiBench: Benchmarking Multi-Stream Multimodal Coordination in Bimanual Embodied Tasks for MLLMs ST-BiBench：用于评估MLLM在双臂操作任务中多模态协调能力的基准测试框架 bi-manual embodied AI large language model
6 CRAFT: Video Diffusion for Bimanual Robot Data Generation CRAFT：利用视频扩散模型生成双臂机器人操作数据，提升泛化性 manipulation bi-manual dual-arm
7 Veo-Act: How Far Can Frontier Video Models Advance Generalizable Robot Manipulation? Veo-Act：探索前沿视频模型在通用机器人操作中的潜力 manipulation dexterous hand physically plausible
8 Sim2Real-AD: A Modular Sim-to-Real Framework for Deploying VLM-Guided Reinforcement Learning in Real-World Autonomous Driving Sim2Real-AD：用于VLM引导强化学习在真实自动驾驶中零样本迁移的模块化框架 sim-to-real sim2real reinforcement learning
9 Drift-Based Policy Optimization: Native One-Step Policy Learning for Online Robot Control 提出基于漂移策略优化(DBPO)的单步生成策略，用于在线机器人控制。 manipulation dual-arm reinforcement learning
10 Watch Your Step: Learning Semantically-Guided Locomotion in Cluttered Environment 提出SemLoco框架，解决腿足机器人复杂环境下的语义引导安全运动问题 legged robot locomotion reinforcement learning
11 From Seeing to Doing: Bridging Reasoning and Decision for Robotic Manipulation 提出FSD模型，通过空间关系推理提升机器人操作的泛化性和零样本性能 manipulation spatial relationship vision-language-action
12 Adaptive Action Chunking at Inference-time for Vision-Language-Action Models 提出自适应动作分块策略，提升视觉-语言-动作模型在机器人操作任务中的性能。 manipulation vision-language-action VLA
13 Steerable Vision-Language-Action Policies for Embodied Reasoning and Hierarchical Control 提出可操纵的视觉-语言-动作策略，用于具身推理和分层控制 manipulation vision-language-action VLA
14 Allometric Scaling Laws for Bipedal Robots 针对双足机器人，提出异速生长比例定律建模方法，优化不同尺寸机器人设计。 legged robot bipedal biped
15 Empowering Multi-Robot Cooperation via Sequential World Models 提出序列世界模型SeqWM，解决多机器人协作中联合动力学建模难题。 quadruped reinforcement learning world model
16 PlayWorld: Learning Robot World Models from Autonomous Play PlayWorld：通过自主玩耍学习机器人世界模型，提升物理交互预测能力 manipulation reinforcement learning world model
17 Low-Cost Teleoperation Extension for Mobile Manipulators 提出一种低成本遥操作方案，用于移动双臂机器人的全身控制 whole-body control manipulation bi-manual
18 Build on Priors: Vision--Language--Guided Neuro-Symbolic Imitation Learning for Data-Efficient Real-World Robot Manipulation 提出基于视觉-语言引导的神经符号模仿学习框架，解决机器人数据高效操作问题。 manipulation imitation learning
19 Embodied-R1: Reinforced Embodied Reasoning for General Robotic Manipulation Embodied-R1：强化具身推理，实现通用机器人操作 manipulation reward design embodied AI
20 Learning-Based Fault Detection for Legged Robots in Remote Dynamic Environments 提出一种基于学习的四足机器人故障检测方法，用于远程动态环境。 quadruped legged robot locomotion
21 A Multi-View 3D Telepresence System for XR Robot Teleoperation 提出多视角3D远程呈现系统，用于XR机器人遥操作，提升操作效率。 manipulation teleoperation
22 FORMULA: FORmation MPC with neUral barrier Learning for safety Assurance FORMULA：融合神经障碍学习的安全多机器人编队MPC MPC model predictive control
23 Learning Geometry-Aware Nonprehensile Pushing and Pulling with Dexterous Hands 提出GD2P，利用灵巧手学习几何感知的非抓取式推拉操作 manipulation dexterous hand
24 SERNF: Sample-Efficient Real-World Dexterous Policy Fine-Tuning via Action-Chunked Critics and Normalizing Flows SERNF：基于动作块评论家和归一化流的高效真实世界灵巧策略微调 manipulation dexterous manipulation multimodal
25 Human-Robot Copilot for Data-Efficient Imitation Learning 提出人机协同框架，通过缩放因子提升模仿学习的数据效率 teleoperation imitation learning
26 Informed Hybrid Zonotope-based Motion Planning Algorithm 提出HZ-MP算法，通过混合zonotope分解和启发式采样解决非凸空间最优路径规划问题。 motion planning
27 Learning Dexterous Grasping from Sparse Taxonomy Guidance GRIT：提出一种基于稀疏分类引导的灵巧抓取学习框架，提升泛化性和可控性。 manipulation dexterous manipulation reinforcement learning
28 Do Robots Need Body Language? Comparing Communication Modalities for Legible Motion Intent in Human-Shared Spaces 研究不同沟通方式对四足机器人运动意图表达的影响，提升人机共享空间安全性。 quadruped multimodal
29 AnyUser: Translating Sketched User Intent into Domestic Robots AnyUser：通过草图将用户意图转化为家用机器人指令 dual-arm multimodal
30 RAPTOR: A Foundation Policy for Quadrotor Control 提出RAPTOR以解决四旋翼控制的适应性问题 sim2real reinforcement learning imitation learning
31 frax: Fast Robot Kinematics and Dynamics in JAX Frax：基于JAX的高性能机器人动力学库，加速机器人控制与学习。 humanoid Unitree
32 Learning to Grasp Anything by Playing with Random Toys 通过玩随机玩具学习抓取：提升机器人操作的泛化能力 manipulation dexterous manipulation
33 Primitive-based Truncated Diffusion for Efficient Trajectory Generation of Differential Drive Mobile Manipulators 提出基于原语截断扩散的运动规划器，高效生成差速移动机械臂轨迹。 trajectory optimization
34 Uncovering Linguistic Fragility in Vision-Language-Action Models via Diversity-Aware Red Teaming 提出多样性感知的红队测试框架DAERT，揭示VLA模型对语言脆弱性 manipulation reinforcement learning embodied AI
35 BiCoord: A Bimanual Manipulation Benchmark towards Long-Horizon Spatial-Temporal Coordination BiCoord：面向长时程空间-时间协同的双臂操作基准 manipulation bi-manual bimanual manipulation ✅
36 ExpressMM: Expressive Mobile Manipulation Behaviors in Human-Robot Interactions ExpressMM：人机交互中基于视觉-语言-动作策略的移动操作表达行为 manipulation mobile manipulation vision-language-action
37 A1: A Fully Transparent Open-Source, Adaptive and Efficient Truncated Vision-Language-Action Model 提出A1框架以降低视觉-语言-动作模型的推理成本 manipulation flow matching affordance
38 Synergizing Efficiency and Reliability for Continuous Mobile Manipulation 提出一种融合效率与可靠性的连续移动操作框架，提升复杂环境下的任务成功率。 manipulation mobile manipulation spatiotemporal
39 JailWAM: Jailbreaking World Action Models in Robot Control 提出JailWAM框架，用于评估和防御世界行为模型在机器人控制中的越狱攻击。 manipulation world action model world action models
40 HiPolicy: Hierarchical Multi-Frequency Action Chunking for Policy Learning HiPolicy：用于策略学习的分层多频动作块划分框架 manipulation policy learning imitation learning
41 Referring-Aware Visuomotor Policy Learning for Closed-Loop Manipulation 提出ReV框架，通过引入参考点增强机器人操作的鲁棒性和动态重规划能力 manipulation policy learning
42 GraspSense: Physically Grounded Grasp and Grip Planning for a Dexterous Robotic Hand via Language-Guided Perception and Force Maps GraspSense：基于力感知的灵巧手抓取与力控规划 humanoid manipulation dexterous manipulation
43 CoEnv: Driving Embodied Multi-Agent Collaboration via Compositional Environment CoEnv：通过组合环境驱动具身多智能体协作 manipulation sim-to-real scene reconstruction
44 Learning-Guided Force-Feedback Model Predictive Control with Obstacle Avoidance for Robotic Deburring 提出一种基于扩散模型引导的力反馈MPC方法，用于机器人去毛刺任务。 MPC model predictive control
45 Precise Aggressive Aerial Maneuvers with Sensorimotor Policies 提出基于强化学习的传感器运动策略，实现无人机在未知环境下高精度、高难度穿越 sim-to-real reinforcement learning policy learning
46 You're Pushing My Buttons: Instrumented Learning of Gentle Button Presses 提出一种基于训练时示教的按钮按压学习方法，降低接触力。 manipulation imitation learning
47 A Co-Design Framework for High-Performance Jumping of a Five-Bar Monoped with Actuator Optimization 提出一种五杆单足跳跃机器人协同设计框架，优化机构、电机、控制参数。 legged robot
48 Simulation-Driven Evolutionary Motion Parameterization for Contact-Rich Granular Scooping with a Soft Conical Robotic Hand 提出基于物理仿真的软锥形手爪抓取运动参数进化优化方法，解决颗粒物料高效抓取问题 trajectory optimization

🔬 支柱九：具身大模型 (Embodied Foundation Models) (10 篇)

#题目一句话要点标签🔗⭐
49 OpenRC: An Open-Source Robotic Colonoscopy Framework for Multimodal Data Acquisition and Autonomy Research OpenRC：用于多模态数据采集和自主研究的开源机器人结肠镜检查框架 vision-language-action VLA multimodal
50 Red-Teaming Vision-Language-Action Models via Quality Diversity Prompt Generation for Robust Robot Policies 提出Q-DIG，利用质量多样性优化进行红队测试，提升VLA机器人策略的鲁棒性。 vision-language-action VLA
51 Precise Robot Command Understanding Using Grammar-Constrained Large Language Models 提出语法约束大语言模型，提升机器人指令理解的精确性 large language model
52 Visual Prompt Based Reasoning for Offroad Mapping using Multimodal LLMs 利用视觉提示的多模态LLM实现越野环境零样本推理与地图构建 multimodal
53 PalpAid: Multimodal Pneumatic Tactile Sensor for Tissue Palpation PalpAid：用于组织触诊的多模态气动触觉传感器，提升机器人辅助手术感知能力 multimodal
54 From Prompt to Physical Action: Structured Backdoor Attacks on LLM-Mediated Robotic Control Systems 揭示LLM机器人控制系统结构性后门攻击漏洞，提出防御方案与安全-响应权衡分析 embodied AI large language model
55 Copilot-Assisted Second-Thought Framework for Brain-to-Robot Hand Motion Decoding 提出基于Copilot辅助的二次思考框架，用于脑-机接口机器人手部运动解码。 multimodal
56 Automating Manual Tasks through Intuitive Robot Programming and Cognitive Robotics 提出一种基于自然交互的机器人编程方法，简化人工任务自动化。 large language model
57 Pre-Execution Safety Gate & Task Safety Contracts for LLM-Controlled Robot Systems SafeGate：面向LLM控制机器人系统的预执行安全门与任务安全契约 large language model
58 AnyImageNav: Any-View Geometry for Precise Last-Meter Image-Goal Navigation AnyImageNav：利用任意视角几何实现精确的末端图像目标导航 foundation model

🔬 支柱二：RL算法与架构 (RL & Architecture) (8 篇)

#题目一句话要点标签🔗⭐
59 Optimizing Neurorobot Policy under Limited Demonstration Data through Preference Regret 提出基于偏好后悔的MYOE框架，解决少样本示教下的神经机器人策略优化问题 reinforcement learning imitation learning SSM
60 ReinVBC: A Model-based Reinforcement Learning Approach to Vehicle Braking Controller 提出ReinVBC，利用离线模型强化学习优化车辆制动控制器 reinforcement learning offline reinforcement learning
61 Fine-tuning is Not Enough: A Parallel Framework for Collaborative Imitation and Reinforcement Learning in End-to-end Autonomous Driving PaIR-Drive：端到端自动驾驶中并行模仿学习与强化学习的协同框架 reinforcement learning imitation learning
62 Diffusion Policy with Bayesian Expert Selection for Active Multi-Target Tracking 提出基于贝叶斯专家选择的扩散策略，用于主动多目标跟踪。 diffusion policy
63 VA-FastNavi-MARL: Real-Time Robot Control with Multimedia-Driven Meta-Reinforcement Learning VA-FastNavi-MARL：基于多媒体驱动的元强化学习实现机器人实时控制 reinforcement learning
64 Towards Safe and Robust Autonomous Vehicle Platooning: A Self-Organizing Cooperative Control Framework 提出TriCoD框架，增强混合交通环境下自动驾驶车队的安全性和鲁棒性 reinforcement learning deep reinforcement learning DRL
65 Grounding Hierarchical Vision-Language-Action Models Through Explicit Language-Action Alignment 提出一种显式语言-动作对齐的层级视觉-语言-动作模型 preference learning vision-language-action VLA
66 Rectified Schrödinger Bridge Matching for Few-Step Visual Navigation 提出RSBM，通过修正Schrödinger桥匹配实现少步视觉导航 flow matching distillation embodied AI

🔬 支柱三：空间感知与语义 (Perception & Semantics) (1 篇)

#题目一句话要点标签🔗⭐
67 Pickalo: Leveraging 6D Pose Estimation for Low-Cost Industrial Bin Picking Pickalo：利用6D位姿估计实现低成本工业分拣 depth estimation 6D pose estimation

🔬 支柱八：物理动画 (Physics-based Animation) (1 篇)

#题目一句话要点标签🔗⭐
68 Temporal Reach-Avoid-Stay Control for Differential Drive Systems via Spatiotemporal Tubes 提出基于时空管道的差速驱动机器人时序可达-避障-停留控制框架 spatiotemporal

🔬 支柱七：动作重定向 (Motion Retargeting) (1 篇)

#题目一句话要点标签🔗⭐
69 Biologically Inspired Event-Based Perception and Sample-Efficient Learning for High-Speed Table Tennis Robots 针对高速乒乓球机器人，提出生物启发式事件感知和高效学习方法 geometric consistency

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
70 Decoupling Torque and Stiffness: A Unified Modeling and Control Framework for Antagonistic Artificial Muscles 针对拮抗式人工肌肉，提出解耦扭矩和刚度的统一建模与控制框架 penetration

⬅️ 返回 cs.RO 首页 · 🏠 返回主页