Dmitry Berenson

University of Michigan

収録論文 68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

• PDDL-ART: デモンストレーションからの自律的シンボリック抽象化による長期的ロボット操作のための視覚言語モデル
  操作計画2026/8/17

  本論文は、単一の専門家デモと自然言語タスク記述からPDDLドメインと問題記述を自動生成するVLMベースのフレームワークPDDL-ARTを提案する。構文・意味・実行レベルの多段階補正パイプラインと、幾何・時間推論のためのツール使用を導入し、長期的操作タスクで有効性を示した。

• 触覚シミュレーションを使わない触覚Sim2Real：ボトルネック潜在再構成による実現
  sim2real2026/8/16

  触覚センサのシミュレーションを避け、シミュレータ内のオラクルセンサ（点群と指先力）で学習したポリシーを、実センサの潜在表現をオラクルに合わせることで実機に転移するフレームワークを提案。挿入や歯車噛み合わせで高成功率を達成。

• 粒子ベース共形予測による接触を考慮した不確実性校正：層状構成空間におけるアプローチ
  不確実性校正2026/8/10

  ロボットの運動予測の不確実性を、接触の有無に応じて局所的に校正する共形予測ベースの手法を提案し、任意の忠実度の粒子モデルを用いて確率的に有効な予測領域を生成する。

• Particle-Based Conformal Prediction for Contact-Aware Uncertainty Calibration in Stratified Configuration Spaces
  2026/8/1
• STEMbot: A Compliant Robot for Under-Canopy Plant Navigation
  2026/7/1
• TAP-VLA: Tactile Annotation Prompting for Vision Language Action Models
  2026/6/1
• PLUME: Probabilistic Latent Unified World Modeling and Parameter Estimation for Multi-Finger Manipulation
  2026/6/1
• Local Conformal Calibration of Dynamics Uncertainty from Semantic Images
  2026/5/1
• Simultaneous Extrinsic Contact and In-Hand Pose Estimation via Distributed Tactile Sensing
  2025/12/1
• Lies We Can Trust: Quantifying Action Uncertainty with Inaccurate Stochastic Dynamics through Conformalized Nonholonomic Lie Groups
  2025/12/1
• Multimodal Diffusion Forcing for Forceful Manipulation
  2025/11/1
• CADRE: Dynamic Catching via Implicit Contact Descriptors and Task-Appropriate Recovery Affordances
  2025/10/1
• Diffusing Trajectory Optimization Problems for Recovery During Multi-Finger Manipulation
  2025/10/1
• AVO: Amortized Value Optimization for Contact Mode Switching in Multi-Finger Manipulation
  2025/10/1
• AnoF-Diff: One-Step Diffusion-Based Anomaly Detection for Forceful Tool Use
  2025/9/1
• Planning-Query-Guided Model Generation for Model-Based Deformable Object Manipulation
  2025/8/1
• Estimating Deformable-Rigid Contact Interactions for a Deformable Tool via Learning and Model-Based Optimization
  2025/5/1
• Language-Guided Object Search in Agricultural Environments
  2025/3/1
• Implicit Contact Diffuser: Sequential Contact Reasoning with Latent Point Cloud Diffusion
  2024/10/1
• Diffusion-Informed Probabilistic Contact Search for Multi-Finger Manipulation
  2024/10/1
• Constraining Gaussian Process Implicit Surfaces for Robot Manipulation via Dataset Refinement
  2024/10/1
• Quantifying Aleatoric and Epistemic Dynamics Uncertainty via Local Conformal Calibration
  2024/9/1
• Multi-finger Manipulation via Trajectory Optimization with Differentiable Rolling and Geometric Constraints
  2024/8/1
• RUMI: Rummaging Using Mutual Information
  2024/8/1
• Tactile-Driven Non-Prehensile Object Manipulation via Extrinsic Contact Mode Control
  2024/5/1
• Online Adaptation of Sampling-Based Motion Planning with Inaccurate Models
  2024/3/1
• Improving Out-of-Distribution Generalization of Learned Dynamics by Learning Pseudometrics and Constraint Manifolds
  2024/3/1
• Subgoal Diffuser: Coarse-to-fine Subgoal Generation to Guide Model Predictive Control for Robot Manipulation
  2024/3/1
• The Grasp Loop Signature: A Topological Representation for Manipulation Planning with Ropes and Cables
  2024/3/1
• Constrained Stein Variational Trajectory Optimization
  2023/8/1
• Motion Planning as Online Learning: A Multi-Armed Bandit Approach to Kinodynamic Sampling-Based Planning
  2023/8/1
• Integrated Object Deformation and Contact Patch Estimation from Visuo-Tactile Feedback
  2023/5/1
• CHSEL: Producing Diverse Plausible Pose Estimates from Contact and Free Space Data
  2023/5/1
• Data-Efficient Learning of Natural Language to Linear Temporal Logic Translators for Robot Task Specification
  2023/3/1
• Learning the Dynamics of Compliant Tool-Environment Interaction for Visuo-Tactile Contact Servoing
  2022/10/1
• Manipulation via Membranes: High-Resolution and Highly Deformable Tactile Sensing and Control
  2022/9/1
• Focused Adaptation of Dynamics Models for Deformable Object Manipulation
  2022/9/1
• Safe Output Feedback Motion Planning from Images via Learned Perception Modules and Contraction Theory
  2022/6/1
• Data Augmentation for Manipulation
  2022/5/1
• Variational Inference MPC using Normalizing Flows and Out-of-Distribution Projection
  2022/5/1
• Soft Tracking Using Contacts for Cluttered Objects to Perform Blind Object Retrieval
  2022/1/1
• Gaussian Process Constraint Learning for Scalable Chance-Constrained Motion Planning from Demonstrations
  2021/12/1
• CLASP: Constrained Latent Shape Projection for Refining Object Shape from Robot Contact
  2021/10/1
• Challenges and Outlook in Robotic Manipulation of Deformable Objects
  2021/5/1
• Model Error Propagation via Learned Contraction Metrics for Safe Feedback Motion Planning of Unknown Systems
  2021/4/1
• Fusing RGBD Tracking and Segmentation Tree Sampling for Multi-Hypothesis Volumetric Segmentation
  2021/4/1
• Keep it Simple: Data-efficient Learning for Controlling Complex Systems with Simple Models
  2021/2/1
• Occlusion-robust Deformable Object Tracking without Physics Simulation
  2021/1/1
• Uncertainty-Aware Constraint Learning for Adaptive Safe Motion Planning from Demonstrations
  2020/11/1
• Tracking Partially-Occluded Deformable Objects while Enforcing Geometric Constraints
  2020/11/1
• Diverse Plausible Shape Completions from Ambiguous Depth Images
  2020/11/1
• TAMPC: A Controller for Escaping Traps in Novel Environments
  2020/10/1
• Planning with Learned Dynamics: Probabilistic Guarantees on Safety and Reachability via Lipschitz Constants
  2020/10/1
• Explaining Multi-stage Tasks by Learning Temporal Logic Formulas from Suboptimal Demonstrations
  2020/6/1
• Inferring Obstacles and Path Validity from Visibility-Constrained Demonstrations
  2020/5/1
• Fast Planning Over Roadmaps via Selective Densification
  2020/2/1
• Manipulating Deformable Objects by Interleaving Prediction, Planning, and Control
  2020/1/1
• Learning Constraints from Locally-Optimal Demonstrations under Cost Function Uncertainty
  2020/1/1
• Learning When to Trust a Dynamics Model for Planning in Reduced State Spaces
  2020/1/1
• Learning Parametric Constraints in High Dimensions from Demonstrations
  2019/10/1
• Robust Humanoid Contact Planning with Learned Zero- and One-Step Capturability Prediction
  2019/9/1
• Inferring Occluded Geometry Improves Performance when Retrieving an Object from Dense Clutter
  2019/7/1
• Learning Constraints from Demonstrations
  2018/12/1
• Efficient Humanoid Contact Planning using Learned Centroidal Dynamics Prediction
  2018/10/1
• Bandit-Based Model Selection for Deformable Object Manipulation
  2017/3/1
• Planning and Resilient Execution of Policies For Manipulation in Contact with Actuation Uncertainty
  2017/3/1
• Goal Set Inverse Optimal Control and Iterative Re-planning for Predicting Human Reaching Motions in Shared Workspaces
  2016/6/1
• Analysis and Observations from the First Amazon Picking Challenge
  2016/1/1