cs.RO（2025-03-04）

📊 共 25 篇论文 | 🔗 5 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (16 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (5 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (3 🔗1) 支柱三：空间感知与语义 (Perception & Semantics) (1)

🔬 支柱一：机器人控制 (Robot Control) (16 篇)

#题目一句话要点标签🔗⭐
1 Learning Dexterous In-Hand Manipulation with Multifingered Hands via Visuomotor Diffusion 提出基于视觉运动扩散策略的多指灵巧手学习框架，实现复杂手内操作 manipulation in-hand manipulation teleoperation
2 Reactive Diffusion Policy: Slow-Fast Visual-Tactile Policy Learning for Contact-Rich Manipulation 提出反应式扩散策略，用于接触丰富操作任务中的视觉-触觉慢-快策略学习。 manipulation teleoperation policy learning
3 Variable-Friction In-Hand Manipulation for Arbitrary Objects via Diffusion-Based Imitation Learning 提出基于扩散模型的模仿学习方法，实现任意物体变摩擦灵巧手操作 manipulation in-hand manipulation imitation learning
4 Natural Selection via Foundation Models for Soft Robot Evolution 提出RoboCrafter-QA基准，利用LLM进行软体机器人进化设计与自然选择。 locomotion manipulation sim-to-real
5 ArticuBot: Learning Universal Articulated Object Manipulation Policy via Large Scale Simulation ArticuBot：通过大规模仿真学习通用铰接物体操作策略 manipulation sim2real motion planning ✅
6 GraphGarment: Learning Garment Dynamics for Bimanual Cloth Manipulation Tasks GraphGarment：学习服装动力学，用于双臂机器人服装操作任务 manipulation bi-manual sim-to-real
7 Accelerating Vision-Language-Action Model Integrated with Action Chunking via Parallel Decoding 提出PD-VLA：一种基于并行解码的视觉-语言-动作模型加速框架 manipulation vision-language-action VLA
8 FalconGym: A Photorealistic Simulation Framework for Zero-Shot Sim-to-Real Vision-Based Quadrotor Navigation FalconGym：用于四旋翼无人机零样本Sim-to-Real视觉导航的逼真模拟框架 sim-to-real imitation learning NeRF
9 Controllable Motion Generation via Diffusion Modal Coupling 提出基于模态耦合扩散模型的可控运动生成方法，提升机器人运动规划和预测精度。 motion planning motion generation
10 Human-aligned Safe Reinforcement Learning for Highway On-Ramp Merging in Dense Traffic 提出一种人机对齐的安全强化学习方法，用于解决高密度交通下的高速公路匝道汇入问题。 MPC model predictive control reinforcement learning
11 Diverse Controllable Diffusion Policy with Signal Temporal Logic 提出基于信号时序逻辑的扩散策略，实现自动驾驶场景下可控、多样且符合规则的智能体行为生成。 trajectory optimization diffusion policy ✅
12 MuBlE: MuJoCo and Blender simulation Environment and Benchmark for Task Planning in Robot Manipulation MuBlE：用于机器人操作任务规划的MuJoCo和Blender仿真环境与基准 manipulation
13 Bridging VLM and KMP: Enabling Fine-grained robotic manipulation via Semantic Keypoints Representation VL-MP：通过语义关键点表示桥接VLM与KMP，实现精细机器人操作 manipulation
14 SEB-Naver: A SE(2)-based Local Navigation Framework for Car-like Robots on Uneven Terrain SEB-Naver：面向崎岖地形类车机器人，基于SE(2)的局部导航框架 trajectory optimization motion planning traversability ✅
15 Predictive Kinematic Coordinate Control for Aerial Manipulators based on Modified Kinematics Learning 提出基于改进运动学学习的预测运动学坐标控制方法，用于提升无人机操作精度。 manipulation MPC model predictive control
16 Scalable Multi-Robot Task Allocation and Coordination under Signal Temporal Logic Specifications 提出基于STL的多机器人任务分配与协调算法，提升复杂约束下大规模系统的可扩展性。 motion planning

🔬 支柱二：RL算法与架构 (RL & Architecture) (5 篇)

#题目一句话要点标签🔗⭐
17 World Models for Anomaly Detection during Model-Based Reinforcement Learning Inference 利用世界模型进行模型预测控制推理过程中的异常检测 reinforcement learning world model
18 A comparison of visual representations for real-world reinforcement learning in the context of vacuum gripping 对比视觉与空间表征，提升真空吸盘抓取任务的强化学习性能 reinforcement learning ✅
19 JPDS-NN: Reinforcement Learning-Based Dynamic Task Allocation for Agricultural Vehicle Routing Optimization 提出JPDS-NN，基于强化学习动态优化农业车辆路径规划，解决不规则田地EDVRP问题。 reinforcement learning
20 Active Robot Curriculum Learning from Online Human Demonstrations 提出基于课程学习的主动机器人示教学习方法，优化在线示教序列。 curriculum learning
21 Learning-Based Passive Fault-Tolerant Control of a Quadrotor with Rotor Failure 提出一种基于学习的四旋翼被动容错控制方法，解决单旋翼故障问题 reinforcement learning behavior cloning

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

#题目一句话要点标签🔗⭐
22 RaceVLA: VLA-based Racing Drone Navigation with Human-like Behaviour RaceVLA：基于VLA的类人行为竞速无人机导航 VLA RT-2 OpenVLA
23 FlowPlan: Zero-Shot Task Planning with LLM Flow Engineering for Robotic Instruction Following 提出FlowPlan以解决机器人指令跟随中的零-shot任务规划问题 instruction following ✅
24 PERCY: Personal Emotional Robotic Conversational System PERCY：一种基于情感感知的个性化机器人对话系统 large language model multimodal

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

#题目一句话要点标签🔗⭐
25 RGBSQGrasp: Inferring Local Superquadric Primitives from Single RGB Image for Graspability-Aware Bin Picking RGBSQGrasp：单目RGB图像超二次曲面抓取，提升无序分拣泛化性 depth estimation metric depth foundation model

⬅️ 返回 cs.RO 首页 · 🏠 返回主页