Saved in:
| Main Authors: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2605.27461 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918525940006912 |
|---|---|
| author | Zhu, Brian Schmitt, Philipp Meister, Philine Gensler, Lukas Khalil, Momen Poggi, Emmanuele Hechtl, Johannes Braunroth, Carsten Wurm, Kai Narayanan, Gokul Solowjow, Eugen von Wichert, Georg Scholz, Andre Albrecht, Felix Metzner, Maxmillian |
| author_facet | Zhu, Brian Schmitt, Philipp Meister, Philine Gensler, Lukas Khalil, Momen Poggi, Emmanuele Hechtl, Johannes Braunroth, Carsten Wurm, Kai Narayanan, Gokul Solowjow, Eugen von Wichert, Georg Scholz, Andre Albrecht, Felix Metzner, Maxmillian |
| contents | Vision-Language-Action (VLA) policies have shown promising manipulation capabilities, yet their practical impact is often limited by the reliability demands of real-world deployment. We present a deployment study of an industrial packaging task at Siemens Factory (GWE, Erlangen, Germany), where a robot must pick a transparent accessory bag from a cluttered pile, insert it into the remaining cavity of a cardboard package, and ensure that the bag and its contents remain below the closing plane. Our goal is to understand the practical effort required to adapt a pretrained Pi0.5 policy to a single factory-floor task through iterative fine-tuning and deployment-driven refinement. The pipeline consists of repeated loops of data collection, curation, fine-tuning, evaluation, and targeted recovery data collection. We have accumulated 2535 episodes (10 hours) from the on-site factory settings. In this paper, we contribute an empirical account of a factory-floor VLA deployment, highlighting recurring failure modes and lessons that inform how to improve the deployment workflow. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_27461 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | A Factory-Floor Deployment Case Study of VLA Pipelines for Industrial Packaging Task: Workflow, Failures, and Lessons Zhu, Brian Schmitt, Philipp Meister, Philine Gensler, Lukas Khalil, Momen Poggi, Emmanuele Hechtl, Johannes Braunroth, Carsten Wurm, Kai Narayanan, Gokul Solowjow, Eugen von Wichert, Georg Scholz, Andre Albrecht, Felix Metzner, Maxmillian Robotics Vision-Language-Action (VLA) policies have shown promising manipulation capabilities, yet their practical impact is often limited by the reliability demands of real-world deployment. We present a deployment study of an industrial packaging task at Siemens Factory (GWE, Erlangen, Germany), where a robot must pick a transparent accessory bag from a cluttered pile, insert it into the remaining cavity of a cardboard package, and ensure that the bag and its contents remain below the closing plane. Our goal is to understand the practical effort required to adapt a pretrained Pi0.5 policy to a single factory-floor task through iterative fine-tuning and deployment-driven refinement. The pipeline consists of repeated loops of data collection, curation, fine-tuning, evaluation, and targeted recovery data collection. We have accumulated 2535 episodes (10 hours) from the on-site factory settings. In this paper, we contribute an empirical account of a factory-floor VLA deployment, highlighting recurring failure modes and lessons that inform how to improve the deployment workflow. |
| title | A Factory-Floor Deployment Case Study of VLA Pipelines for Industrial Packaging Task: Workflow, Failures, and Lessons |
| topic | Robotics |
| url | https://arxiv.org/abs/2605.27461 |