Florian T. Pokorny
KTH Royal Institute of Technology
収録論文 37本 ・ フィジカルAI/ロボット学習
視覚的注意/ポリシー学習
※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
- 行動から注意を学ぶ：ポリシー学習のための創発的視覚ボトルネック視覚的注意/ポリシー学習2026/8/13
行動教師信号のみから注目領域（ROI）を学習するSeekerを提案し、RGBクロップや点群フィルタリングの空間インターフェースとして利用。実機で成功率を大幅に向上させた。
- Attention from Action, for Action: Emergent Visual Bottlenecks for Policy Learning2026/8/1
- MirrorDuo: Reflection-Consistent Visuomotor Learning from Mirrored Demonstration Pairs2026/6/1
- Robustness of Robotic Manipulation: Foundations and Frontiers2026/6/1
- Physics-Informed Eikonal Caging for Whole-Arm Manipulation Planning2026/6/1
- Fibration Trees: A Unified Approach to Multi-Robot Motion Planning2026/6/1
- Adversarial Attacks on Learned Policies for Surgical Robotic Tasks2026/6/1
- MoDex: A Diffusion Policy for Sequential Multi-Object Dexterous Grasping2026/6/1
- Speculative Policy Orchestration: A Latency-Resilient Framework for Cloud-Robotic Manipulation2026/3/1
- RoboLight: A Dataset with Linearly Composable Illumination for Robotic Manipulation2026/3/1
- Latent Diffeomorphic Co-Design of End-Effectors for Deformable and Fragile Object Manipulation2026/2/1
- PALM: Enhanced Generalizability for Local Visuomotor Policies via Perception Alignment2026/1/1
- ViTA-Seg: Vision Transformer for Amodal Segmentation in Robotics2025/12/1
- Sim-to-Real Gentle Manipulation of Deformable and Fragile Objects with Stress-Guided Reinforcement Learning2025/10/1
- Physically-based Lighting Generation for Robotic Manipulation2025/8/1
- Robustness-Aware Tool Selection and Manipulation Planning with Learned Energy-Informed Guidance2025/6/1
- Unifying Complementarity Constraints and Control Barrier Functions for Safe Whole-Body Robot Control2025/4/1
- Data-augmented Learning of Geodesic Distances in Irregular Domains through Soner Boundary Conditions2025/3/1
- R900: Understanding the Cost-Effectiveness of Random Exploration from 900 Hours of Robotic Data Collection2025/3/1
- Grasping a Handful: Sequential Multi-Object Dexterous Grasp Generation2025/3/1
- T-DOM: A Taxonomy for Robotic Manipulation of Deformable Objects2024/12/1
- PACA: Perspective-Aware Cross-Attention Representation for Zero-Shot Scene Rearrangement2024/10/1
- Feature Extractor or Decision Maker: Rethinking the Role of Visual Encoders in Visuomotor Policies2024/9/1
- CageCoOpt: Enhancing Manipulation Robustness through Caging-Guided Morphology and Policy Co-Optimization2024/9/1
- Forward Invariance in Trajectory Spaces for Safety-critical Control2024/7/1
- Adaptive Distance Functions via Kelvin Transformation2024/6/1
- Walk on Spheres for PDE-based Path Planning2024/6/1
- Characterizing Manipulation Robustness through Energy Margin and Caging Analysis2024/4/1
- How Physics and Background Attributes Impact Video Transformers in Robotic Manipulation: A Case Study on Planar Pushing2023/10/1
- CloudGripper: An Open Source Cloud Robotics Testbed for Robotic Manipulation Research, Benchmarking and Data Collection at Scale2023/9/1
- Quasi-static Soft Fixture Analysis of Rigid and Deformable Objects2023/9/1
- BITKOMO: Combining Sampling and Optimization for Fast Convergence in Optimal Motion Planning2022/3/1
- Approximate Topological Optimization using Multi-Mode Estimation for Robot Motion Planning2021/7/1
- Free Space of Rigid Objects: Caging, Path Non-Existence, and Narrow Passage Detection2020/2/1
- A Decomposition-Based Approach to Reasoning about Free Space Path-Connectivity for Rigid Objects in 2D2017/10/1
- CapriDB - Capture, Print, Innovate: A Low-Cost Pipeline and Database for Reproducible Manipulation Research2016/10/1
- HIRL: Hierarchical Inverse Reinforcement Learning for Long-Horizon Tasks with Delayed Rewards2016/4/1