Physics-Aware Robotic Palletization with Online Masking Inference
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_ | 1866909502125637632 |
|---|---|
| author | Zhang, Tianqi Wu, Zheng Chen, Yuxin Wang, Yixiao Liang, Boyuan Moura, Scott Tomizuka, Masayoshi Ding, Mingyu Zhan, Wei |
| author_facet | Zhang, Tianqi Wu, Zheng Chen, Yuxin Wang, Yixiao Liang, Boyuan Moura, Scott Tomizuka, Masayoshi Ding, Mingyu Zhan, Wei |
| contents | The efficient planning of stacking boxes, especially in the online setting where the sequence of item arrivals is unpredictable, remains a critical challenge in modern warehouse and logistics management. Existing solutions often address box size variations, but overlook their intrinsic and physical properties, such as density and rigidity, which are crucial for real-world applications. We use reinforcement learning (RL) to solve this problem by employing action space masking to direct the RL policy toward valid actions. Unlike previous methods that rely on heuristic stability assessments which are difficult to assess in physical scenarios, our framework utilizes online learning to dynamically train the action space mask, eliminating the need for manual heuristic design. Extensive experiments demonstrate that our proposed method outperforms existing state-of-the-arts. Furthermore, we deploy our learned task planner in a real-world robotic palletizer, validating its practical applicability in operational settings. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_13443 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Physics-Aware Robotic Palletization with Online Masking Inference Zhang, Tianqi Wu, Zheng Chen, Yuxin Wang, Yixiao Liang, Boyuan Moura, Scott Tomizuka, Masayoshi Ding, Mingyu Zhan, Wei Robotics The efficient planning of stacking boxes, especially in the online setting where the sequence of item arrivals is unpredictable, remains a critical challenge in modern warehouse and logistics management. Existing solutions often address box size variations, but overlook their intrinsic and physical properties, such as density and rigidity, which are crucial for real-world applications. We use reinforcement learning (RL) to solve this problem by employing action space masking to direct the RL policy toward valid actions. Unlike previous methods that rely on heuristic stability assessments which are difficult to assess in physical scenarios, our framework utilizes online learning to dynamically train the action space mask, eliminating the need for manual heuristic design. Extensive experiments demonstrate that our proposed method outperforms existing state-of-the-arts. Furthermore, we deploy our learned task planner in a real-world robotic palletizer, validating its practical applicability in operational settings. |
| title | Physics-Aware Robotic Palletization with Online Masking Inference |
| topic | Robotics |
| url | https://arxiv.org/abs/2502.13443 |