cs.RO（2026-03-19）

📊 共 32 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (26 🔗4) 支柱九：具身大模型 (Embodied Foundation Models) (3) 支柱三：空间感知与语义 (Perception & Semantics) (3)

🔬 支柱一：机器人控制 (Robot Control) (26 篇)

#题目一句话要点标签🔗⭐
1 Empathetic Motion Generation for Humanoid Educational Robots via Reasoning-Guided Vision--Language--Motion Diffusion Architecture 提出RG-VLMD框架，为教育机器人生成具身同语姿势，提升教学表达力。 humanoid humanoid robot motion diffusion
2 Graph-of-Constraints Model Predictive Control for Reactive Multi-agent Task and Motion Planning 提出GoC-MPC，解决多智能体任务与运动规划中动态约束难题 manipulation MPC model predictive control
3 ADMM-Based Distributed MPC with Control Barrier Functions for Safe Multi-Robot Quadrupedal Locomotion 提出基于ADMM和CBF的分布式MPC框架，用于安全的多足机器人运动控制。 quadruped whole-body control locomotion
4 Scaling Sim-to-Real Reinforcement Learning for Robot VLAs with Generative 3D Worlds 利用生成式3D世界，扩展机器人VLA的Sim-to-Real强化学习 sim-to-real domain randomization reinforcement learning
5 PRIOR: Perceptive Learning for Humanoid Locomotion with Reference Gait Priors PRIOR：基于参考步态先验的人形机器人感知运动学习框架 humanoid humanoid locomotion locomotion
6 CSSDF-Net: Safe Motion Planning Based on Neural Implicit Representations of Configuration Space Distance Field 提出CSSDF-Net，通过构型空间距离场的神经隐式表示实现安全运动规划。 MPC trajectory optimization motion planning
7 V-Dreamer: Automating Robotic Simulation and Trajectory Synthesis via Video Generation Priors V-Dreamer：利用视频生成先验自动生成机器人仿真环境与轨迹 manipulation sim-to-real imitation learning
8 Not All Features Are Created Equal: A Mechanistic Study of Vision-Language-Action Models 揭示VLA模型工作机制：视觉主导动作生成，语言敏感性依赖任务结构。 manipulation vision-language-action VLA
9 Sparse Autoencoders Reveal Interpretable and Steerable Features in VLA Models 稀疏自编码器揭示VLA模型中可解释和可控的特征，提升机器人操作泛化性 manipulation vision-language-action VLA
10 OmniVTA: Visuo-Tactile World Modeling for Contact-Rich Robotic Manipulation OmniVTA：用于接触式机器人操作的视觉-触觉世界建模框架 manipulation world model contact-aware ✅
11 Articulated-Body Dynamics Network: Dynamics-Grounded Prior for Robot Learning 提出基于动力学先验的关节体动力学网络，提升机器人学习效率与泛化性 quadruped humanoid locomotion
12 Safety-Guaranteed Imitation Learning from Nonlinear Model Predictive Control for Spacecraft Close Proximity Operations 提出一种基于模仿学习和控制屏障函数的航天器近距离操作安全控制框架 model predictive control imitation learning
13 ViTac-Tracing: Visual-Tactile Imitation Learning of Deformable Object Tracing ViTac-Tracing：提出视觉-触觉模仿学习方法，解决柔性物体追踪问题。 manipulation sim-to-real teleoperation
14 Efficient and Versatile Quadrupedal Skating: Optimal Co-design via Reinforcement Learning and Bayesian Optimization 基于强化学习与贝叶斯优化的四足机器人滑冰高效协同设计 quadruped reinforcement learning
15 Lightweight Model Predictive Control for Spacecraft Rendezvous Attitude Synchronization 针对航天器交会对接姿态同步，提出轻量化模型预测控制方法 MPC model predictive control
16 Robotic Agentic Platform for Intelligent Electric Vehicle Disassembly RAPID平台：用于智能电动汽车拆解的机器人代理平台 manipulation open-vocabulary open vocabulary
17 Computationally Efficient Density-Driven Optimal Control via Analytical KKT Reduction and Contractive MPC 提出基于解析KKT简化的密度驱动最优控制，实现多智能体高效空间分布 MPC
18 MemoAct: Atkinson-Shiffrin-Inspired Memory-Augmented Visuomotor Policy for Robotic Manipulation MemoAct：受Atkinson-Shiffrin模型启发的记忆增强型机器人操作策略 manipulation ✅
19 Introducing M: A Modular, Modifiable Social Robot M：一种模块化、可修改的社交机器人平台，降低研究门槛。 sim-to-real multimodal
20 Fire as a Service: Augmenting Robot Simulators with Thermally and Visually Accurate Fire Dynamics FaaS：通过热力学和视觉精确的火焰动力学增强机器人仿真 teleoperation multimodal
21 Contact Status Recognition and Slip Detection with a Bio-inspired Tactile Hand 提出一种基于仿生触觉手的接触状态识别与滑移检测方法，适用于易碎物体操作。 manipulation multimodal
22 ATG-MoE: Autoregressive trajectory generation with mixture-of-experts for assembly skill learning 提出ATG-MoE，用于解决机器人装配技能学习中的泛化性与多技能集成问题。 manipulation ✅
23 SOFTMAP: Sim2Real Soft Robot Forward Modeling via Topological Mesh Alignment and Physics Prior 提出SOFTMAP以解决软机器人前向建模问题 sim-to-real sim2real teleoperation
24 Speculative Policy Orchestration: A Latency-Resilient Framework for Cloud-Robotic Manipulation 提出推测策略编排SPO，解决云机器人操作中网络延迟导致的控制问题。 manipulation motion planning world model
25 Unlabeled Multi-Robot Motion Planning with Improved Separation Trade-offs 针对密集环境，提出改进分离权衡的无标签多机器人运动规划算法 motion planning
26 VAMPO: Policy Optimization for Improving Visual Dynamics in Video Action Models 提出VAMPO以优化视频动作模型中的视觉动态问题 manipulation ✅

🔬 支柱九：具身大模型 (Embodied Foundation Models) (3 篇)

#题目一句话要点标签🔗⭐
27 FASTER: Rethinking Real-Time Flow VLAs FASTER：重新思考实时Flow VLA，显著降低反应延迟 vision-language-action VLA
28 NavTrust: Benchmarking Trustworthiness for Embodied Navigation NavTrust：构建具身导航可信度评测基准，评估真实场景下的鲁棒性 VLN
29 Can LLMs Prove Robotic Path Planning Optimality? A Benchmark for Research-Level Algorithm Verification 构建机器人路径规划最优性证明基准，评估LLM在研究级算法验证中的能力 large language model chain-of-thought

🔬 支柱三：空间感知与语义 (Perception & Semantics) (3 篇)

#题目一句话要点标签🔗⭐
30 REST: Receding Horizon Explorative Steiner Tree for Zero-Shot Object-Goal Navigation 提出REST框架，通过探索性Steiner树优化零样本物体目标导航中的路径规划。 scene understanding open-vocabulary open vocabulary
31 Benchmarking Visual Feature Representations for LiDAR-Inertial-Visual Odometry Under Challenging Conditions 提出混合视觉特征的LIVO框架，提升恶劣光照条件下的定位精度 visual odometry feature matching
32 ROFT-VINS: Robust Feature Tracking-based Visual-Inertial State Estimation for Harsh Environment 提出基于深度学习的鲁棒特征跟踪方法，提升VIO在恶劣环境下的定位精度 VIO

⬅️ 返回 cs.RO 首页 · 🏠 返回主页