Weiming Zhi

The University of Sydney

収録論文 47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

• NestDex: コパイロット支援遠隔操作による入れ子型ポリシー学習を用いた器用な操作
  マニピュレーション2026/8/13

  器用な操作のデモ収集を容易にするため、学習済みの手先スキルを遠隔操作に組み込んだ入れ子型ポリシー学習フレームワークを提案し、実機実験で有効性を示した。

• 接触前実行監視：行動条件付き潜在世界モデルによるコンタクトガード
  マニピュレーション2026/8/13

  ロボットの接触を伴う操作において、接触前に失敗を予測して動作を中断する監視システムを提案。潜在空間での世界モデルを用いて、計画された行動の結果を予測し、失敗の可能性を検出する。

• AutoIntervene: アクションチャンキング模倣学習ポリシーのための校正された介入
  模倣学習/介入2026/8/7

  アクションチャンキング模倣学習ポリシーがデモ分布から逸脱した際に、オペレータへの制御移行を選択的に行うオンラインフレームワークを提案。視覚・行動サポートメモリと校正された閾値で介入を制御し、実世界の双腕操作タスクで成功率を向上させた。

• AutoIntervene: Calibrated Intervention for Action-Chunking Imitation Learning Policies
  2026/8/1
• ContactGuard: Pre-Contact Execution Monitoring with Action-Conditioned Latent World Models
  2026/8/1
• NestDex: Nested Policy Learning with Copilot Assisted Teleoperation for Dexterous Manipulation
  2026/8/1
• Tri-Manual Visuomotor Imitation Learning of Robot Policies
  2026/7/1
• SemAnCorr: Semantic Anchored Correspondence for Zero-Shot Manipulation Skill Transfer
  2026/7/1
• TRACE: Trajectory-Routed Causal Memory for Delayed-Evidence Visuomotor Imitation
  2026/6/1
• Robots that Collaborate: Sequential Asymmetric Imitation for Learning Coupled Robot Policies
  2026/6/1
• PATCH: Action-Chunk-Conditioned Latent Patch Innovation Monitoring for Robot Manipulation
  2026/6/1
• BayesFP: Posterior Estimation for Flow-Based Policies via Feynman-Kac Sampling
  2026/6/1
• Rewind-IL: Online Failure Detection and State Respawning for Imitation Learning
  2026/4/1
• SOFTMAP: Sim2Real Soft Robot Forward Modeling via Topological Mesh Alignment and Physics Prior
  2026/3/1
• ContraMap: Contrastive Uncertainty Mapping for Robot Environment Representation
  2026/3/1
• Constraining Streaming Flow Models for Adapting Learned Robot Trajectory Distributions
  2026/2/1
• TriPilot-FF: Coordinated Whole-Body Teleoperation with Force Feedback
  2026/2/1
• Affordance Transfer Across Object Instances via Semantically Anchored Functional Map
  2026/2/1
• Cross-Modal Instructions for Robot Motion Generation
  2025/9/1
• ConEQsA: Concurrent and Asynchronous Embodied Questions Scheduling and Answering
  2025/9/1
• Joint Flow Trajectory Optimization For Feasible Robot Motion Generation from Video Demonstrations
  2025/9/1
• Efficient Construction of Implicit Surface Models From a Single Image for Motion Generation
  2025/9/1
• Bi-Manual Joint Camera Calibration and Scene Representation
  2025/5/1
• From Single Images to Motion Policies via Video-Generation Environment Representations
  2025/5/1
• GraphSeg: Segmented 3D Representations via Graph Edge Addition and Contraction
  2025/4/1
• Infinite Leagues Under the Sea: Photorealistic 3D Underwater Terrain Generation by Latent Fractal Diffusion Models
  2025/3/1
• Robust Bayesian Scene Reconstruction with Retrieval-Augmented Priors for Precise Grasping and Planning
  2024/11/1
• PhotoReg: Photometrically Registering 3D Gaussian Splatting Models
  2024/10/1
• SplaTraj: Camera Trajectory Generation with Semantic Gaussian Splatting
  2024/10/1
• ModCube: Modular, Self-Assembling Cubic Underwater Robot
  2024/9/1
• Learning to Represent Surroundings, Anticipate Motion and Take Informed Actions in Unstructured Environments
  2024/7/1
• 3D Foundation Models Enable Simultaneous Geometry and Pose Estimation of Grasped Objects
  2024/7/1
• Unifying Scene Representation and Hand-Eye Calibration with 3D Foundation Models
  2024/4/1
• Teaching Robots Where To Go And How To Act With Human Sketches via Spatial Diagrammatic Instructions
  2024/3/1
• DarkGS: Learning Neural Illumination and 3D Gaussians Relighting for Robotic Exploration in the Dark
  2024/3/1
• V-PRISM: Probabilistic Mapping of Unknown Tabletop Scenes
  2024/3/1
• Instructing Robots by Sketching: Learning from Demonstration via Probabilistic Diagrammatic Teaching
  2023/9/1
• Learning Orbitally Stable Systems for Diagrammatically Teaching
  2023/9/1
• Pedestrian Trajectory Prediction Using Dynamics-based Deep Learning
  2023/9/1
• Multi-query Robotic Manipulator Task Sequencing with Gromov-Hausdorff Approximations
  2022/9/1
• L4KDE: Learning for KinoDynamic Tree Expansion
  2022/3/1
• Parallelised Diffeomorphic Sampling-based Motion Planning
  2021/8/1
• Probabilistic Trajectory Prediction with Structural Constraints
  2021/7/1
• Anticipatory Navigation in Crowds by Probabilistic Prediction of Pedestrian Future Movements
  2020/11/1
• Occ-Traj120: Occupancy Maps with Associated Trajectories
  2019/9/1
• OCTNet: Trajectory Generation in New Environments from Past Experiences
  2019/9/1
• Kernel Trajectory Maps for Multi-Modal Probabilistic Motion Prediction
  2019/7/1