cs.RO（2025-04-09）

📊 共 19 篇论文

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (10) 支柱三：空间感知与语义 (Perception & Semantics) (4) 支柱二：RL算法与架构 (RL & Architecture) (4) 支柱九：具身大模型 (Embodied Foundation Models) (1)

🔬 支柱一：机器人控制 (Robot Control) (10 篇)

#题目一句话要点标签🔗⭐
1 Sim-to-Real of Humanoid Locomotion Policies via Joint Torque Space Perturbation Injection 提出基于关节力矩空间扰动注入的Sim-to-Real人形机器人运动策略 legged robot humanoid humanoid locomotion
2 RAMBO: RL-Augmented Model-Based Whole-Body Control for Loco-Manipulation RAMBO：强化学习增强的模型预测控制用于腿式机器人灵巧操作 quadruped legged robot bipedal
3 Bridging Deep Reinforcement Learning and Motion Planning for Model-Free Navigation in Cluttered Environments 融合运动规划与深度强化学习，解决复杂环境下的模型无关导航问题 motion planning reinforcement learning deep reinforcement learning
4 OPAL: Encoding Causal Understanding of Physical Systems for Robot Learning OPAL：通过拓扑约束增强机器人学习的物理系统因果理解 manipulation flow matching vision-language-action
5 Collision avoidance from monocular vision trained with novel view synthesis 提出基于单目视觉和新视角合成的避障策略，无需显式环境建模。 locomotion locomotion policy gaussian splatting
6 Two by Two: Learning Multi-Task Pairwise Objects Assembly for Generalizable Robot Manipulation 提出2BY2数据集和基于等变特征的两步SE(3)位姿估计方法，用于通用机器人操作中的多任务成对物体组装。 manipulation spatial relationship
7 Leveraging GCN-based Action Recognition for Teleoperation in Daily Activity Assistance 提出基于GCN动作识别的遥操作框架，用于日常活动辅助 motion planning teleoperation
8 Developing Modular Grasping and Manipulation Pipeline Infrastructure to Streamline Performance Benchmarking 构建模块化抓取与操作流水线，简化性能基准测试 manipulation
9 Overcoming Dynamic Environments: A Hybrid Approach to Motion Planning for Manipulators 提出混合运动规划方法，提升机械臂在动态环境中的适应性和效率 motion planning
10 Data-Enabled Neighboring Extremal: Case Study on Model-Free Trajectory Tracking for Robotic Arm 提出数据驱动邻域极值（DeeNE）框架，加速机器人手臂模型无关轨迹跟踪。 model predictive control

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

#题目一句话要点标签🔗⭐
11 RayFronts: Open-Set Semantic Ray Frontiers for Online Scene Understanding and Exploration RayFronts：用于在线场景理解和探索的开放集语义射线前沿 scene understanding semantic mapping semantic map
12 Domain-Conditioned Scene Graphs for State-Grounded Task Planning 提出域条件场景图，提升机器人任务规划中状态表示的准确性和规划成功率 scene understanding multimodal
13 Towards Efficient Roadside LiDAR Deployment: A Fast Surrogate Metric Based on Entropy-Guided Visibility 提出基于熵引导可见性的快速代理指标，用于优化路侧LiDAR部署。 occupancy grid
14 ASHiTA: Automatic Scene-grounded HIerarchical Task Analysis 提出ASHiTA框架以解决高层任务分析与3D场景关联问题 scene reconstruction

🔬 支柱二：RL算法与架构 (RL & Architecture) (4 篇)

#题目一句话要点标签🔗⭐
15 Interactive Expressive Motion Generation Using Dynamic Movement Primitives 利用动态运动基元实现交互式、富有表现力的机器人动作生成 imitation learning motion generation
16 Dynamic Residual Safe Reinforcement Learning for Multi-Agent Safety-Critical Scenarios Decision-Making 提出动态残差安全强化学习框架，解决多智能体安全决策问题 reinforcement learning spatiotemporal
17 Learning global control of underactuated systems with Model-Based Reinforcement Learning 提出基于模型的强化学习方法MC-PILCO，用于欠驱动系统的全局控制学习。 reinforcement learning
18 Controller Distillation Reduces Fragile Brain-Body Co-Adaptation and Enables Migrations in MAP-Elites 提出控制器蒸馏方法，解决MAP-Elites中脑-体协同适应脆弱性问题，提升迁移能力。 distillation

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

#题目一句话要点标签🔗⭐
19 A Multi-Modal Interaction Framework for Efficient Human-Robot Collaborative Shelf Picking 提出多模态交互框架，提升人机协作货架拣选效率 large language model multimodal chain-of-thought

⬅️ 返回 cs.RO 首页 · 🏠 返回主页