Λεπτομέρειες βιβλιογραφικής εγγραφής
| Τίτλος: |
Assembly task planning framework based on knowledge graph. |
| Συγγραφείς: |
Xu, Zhaobo, Zhang, Chaoran, Hou, Sheng, Han, Zhaochun, Zeng, Long, Feng, Pingfa |
| Πηγή: |
Journal of Intelligent Manufacturing; Sep2026, Vol. 37 Issue 9, p3481-3504, 24p |
| Θεματικοί όροι: |
Knowledge graphs, Flexible manufacturing systems, Work design, Industrial robots, Program generators (Computer programs), Task analysis |
| Περίληψη: |
Assembly task planning (ATP) translates natural language described assembly tasks into executable programs, along with desired product models as input. It is a critical component of embodied intelligent assembly systems. However, current methods leveraging large language models are unable to comply with industrial standards and constraints well. We address this issue by proposing a knowledge graph (KG) based ATP framework that comprises four key modules: assembly KG, high-level task planner, low-level skill controller and reconfigurable flexible assembly system. First, the information of a product family is stored in an assembly KG. Then, a given language described assembly task is decomposed into a sequence of subtasks by the high-level task planner based on product information and inference rules in the assembly KG. After that, the low-level skill controller further transforms the assembly subtasks into skills that can be ultimately mapped to assembly programs, which can be directly executed on our highly reconfigurable flexible assembly system. Moreover, we develop a software platform to ease the assembly program generation. The proposed method is demonstrated by its application to the assembly of four valves. The results demonstrate that the programming time has been reduced from 10 to 25 min to less than 1 s. And the assembly accuracy can reach over 80%. These show the effectiveness and potential of our KG-based ATP framework in improving efficiency and reducing manual workload in industrial assembly scenarios. [ABSTRACT FROM AUTHOR] |
|
Copyright of Journal of Intelligent Manufacturing is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) |
| Βάση Δεδομένων: |
Complementary Index |