Histogram Transporter: Learning Rotation-Equivariant Orientation Histograms for High-Precision Robotic Kitting
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910877882515456 |
|---|---|
| author | Zhou, Jiadong Zeng, Yadan Dong, Huixu Chen, I-Ming |
| author_facet | Zhou, Jiadong Zeng, Yadan Dong, Huixu Chen, I-Ming |
| contents | Robotic kitting is a critical task in industrial automation that requires the precise arrangement of objects into kits to support downstream production processes. However, when handling complex kitting tasks that involve fine-grained orientation alignment, existing approaches often suffer from limited accuracy and computational efficiency. To address these challenges, we propose Histogram Transporter, a novel kitting framework that learns high-precision pick-and-place actions from scratch using only a few demonstrations. First, our method extracts rotation-equivariant orientation histograms (EOHs) from visual observations using an efficient Fourier-based discretization strategy. These EOHs serve a dual purpose: improving picking efficiency by directly modeling action success probabilities over high-resolution orientations and enhancing placing accuracy by serving as local, discriminative feature descriptors for object-to-placement matching. Second, we introduce a subgroup alignment strategy in the place model that compresses the full spectrum of EOHs into a compact orientation representation, enabling efficient feature matching while preserving accuracy. Finally, we examine the proposed framework on the simulated Hand-Tool Kitting Dataset (HTKD), where it outperforms competitive baselines in both success rates and computational efficiency. Further experiments on five Raven-10 tasks exhibits the remarkable adaptability of our approach, with real-robot trials confirming its applicability for real-world deployment. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_12541 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Histogram Transporter: Learning Rotation-Equivariant Orientation Histograms for High-Precision Robotic Kitting Zhou, Jiadong Zeng, Yadan Dong, Huixu Chen, I-Ming Robotics Robotic kitting is a critical task in industrial automation that requires the precise arrangement of objects into kits to support downstream production processes. However, when handling complex kitting tasks that involve fine-grained orientation alignment, existing approaches often suffer from limited accuracy and computational efficiency. To address these challenges, we propose Histogram Transporter, a novel kitting framework that learns high-precision pick-and-place actions from scratch using only a few demonstrations. First, our method extracts rotation-equivariant orientation histograms (EOHs) from visual observations using an efficient Fourier-based discretization strategy. These EOHs serve a dual purpose: improving picking efficiency by directly modeling action success probabilities over high-resolution orientations and enhancing placing accuracy by serving as local, discriminative feature descriptors for object-to-placement matching. Second, we introduce a subgroup alignment strategy in the place model that compresses the full spectrum of EOHs into a compact orientation representation, enabling efficient feature matching while preserving accuracy. Finally, we examine the proposed framework on the simulated Hand-Tool Kitting Dataset (HTKD), where it outperforms competitive baselines in both success rates and computational efficiency. Further experiments on five Raven-10 tasks exhibits the remarkable adaptability of our approach, with real-robot trials confirming its applicability for real-world deployment. |
| title | Histogram Transporter: Learning Rotation-Equivariant Orientation Histograms for High-Precision Robotic Kitting |
| topic | Robotics |
| url | https://arxiv.org/abs/2503.12541 |