Heng Zhang

Istituto Italiano di Tecnologia

収録論文 36本 ・ フィジカルAI/ロボット学習

※arXiv著者名で収集。同姓同名の別人の論文が含まれる場合があります。

論文

• 集中訓練と批評家分解による実世界オンライン強化学習の効率化：ロボット操作への応用
  マニピュレーション2026/8/10

  実世界でのオンライン強化学習を効率化するため、集中訓練と分散実行（CTDE）とハイブリッド報酬アーキテクチャ（HRA）を組み合わせ、複数のエージェントが共有する多頭批評家を導入した。タスク報酬と把握報酬を分離し、サンプル効率と性能を大幅に向上させた。

• Efficient Real-World Online Reinforcement Learning for Robot Manipulation via Centralized Training and Critic Decomposition
  2026/8/1
• MANGO-Grasp: Mahalanobis Fields over Geometry-Oriented 3D Gaussians for Cross-Embodiment Dexterous Grasping
  2026/8/1
• AnchorVLA: Bridging Discrete Decisions and Continuous Trajectories for Vision-Language-Action Planning
  2026/7/1
• Agentic Real2Sim: Physics-based World Modeling with Vision-Language Agents
  2026/7/1
• BoxTwin: Learning Elastoplastic Articulated Object Dynamics from Videos
  2026/7/1
• NeuralActuator: Neural Actuation Modeling for Robot Dynamics and External Force Perception
  2026/7/1
• HiL-ResRL: A Model-Agnostic Finetuning Adapter via Human-in-the-loop Residual Reinforcement Learning
  2026/6/1
• UniLab: A Heterogeneous Architecture for Robot RL Beyond GPU-Dominant Paradigms
  2026/5/1
• HyperSim: A Holistic Sim-To-Real Framework For Robust Robotic Manipulation
  2026/5/1
• Learning Tactile-Aware Quadrupedal Loco-Manipulation Policies
  2026/4/1
• GS-Playground: A High-Throughput Photorealistic Simulator for Vision-Informed Robot Learning
  2026/4/1
• TacVLA: Contact-Aware Tactile Fusion for Robust Vision-Language-Action Manipulation
  2026/3/1
• Self-supervised Physics-Informed Manipulation of Deformable Linear Objects with Non-negligible Dynamics
  2026/2/1
• Semantic-Contact Fields for Category-Level Generalizable Tactile Tool Manipulation
  2026/2/1
• PLanAR: Planning-Language-Grounded Agentic Reasoning for Robot Manipulation
  2026/2/1
• CompliantVLA-adaptor: VLM-Guided Variable Impedance Action for Safe Contact-Rich Manipulation
  2026/1/1
• Safe Learning for Contact-Rich Robot Tasks: A Survey from Classical Learning-Based Methods to Safe Foundation Models
  2025/12/1
• OmniVIC: A Self-Improving Variable Impedance Controller with Vision-Language In-Context Learning for Safe Robotic Manipulation
  2025/10/1
• MachaGrasp: Morphology-Aware Cross-Embodiment Dexterous Hand Articulation Generation for Grasping
  2025/10/1
• ActivePose: Active 6D Object Pose Estimation and Tracking for Robotic Manipulation
  2025/9/1
• AsyMoE: Leveraging Modal Asymmetry for Enhanced Expert Specialization in Large Vision-Language Models
  2025/9/1
• INTENTION: Inferring Tendencies of Humanoid Robot Motion Through Interactive Intuition and Grounded VLM
  2025/8/1
• HiBerNAC: Hierarchical Brain-emulated Robotic Neural Agent Collective for Disentangling Complex Manipulation
  2025/6/1
• A Survey on Imitation Learning for Contact-Rich Tasks in Robotics
  2025/6/1
• Scaling Laws in Scientific Discovery with AI and Robot Scientists
  2025/3/1
• TERL: Large-Scale Multi-Target Encirclement Using Transformer-Enhanced Reinforcement Learning
  2025/3/1
• Passivity-Centric Safe Reinforcement Learning for Contact-Rich Robotic Tasks
  2025/3/1
• Bresa: Bio-inspired Reflexive Safe Reinforcement Learning for Contact-Rich Robotic Tasks
  2025/3/1
• SRL-VIC: A Variable Stiffness-Based Safe Reinforcement Learning for Contact-Rich Robotic Tasks
  2024/6/1
• An Intraoperative Force Perception and Signal Decoupling Method on Capsulorhexis Forceps
  2023/11/1
• Refining 6-DoF Grasps with Context-Specific Classifiers
  2023/8/1
• A Novel Graph-based Motion Planner of Multi-Mobile Robot Systems with Formation and Obstacle Constraints
  2022/10/1
• A Novel Semi-Coupled Hierarchical Motion Planning Framework for Cooperative Transportation of Multiple Mobile Manipulators
  2022/8/1
• Multi-Robot Object Transport Motion Planning with a Deformable Sheet
  2021/11/1
• Design and Control of a Highly Redundant Rigid-Flexible Coupling Robot to Assist the COVID-19 Oropharyngeal-Swab Sampling
  2021/2/1