Fully Automated Cyclic Planning for Large-Scale Manufacturing Domains

Authors

  • Masataro Asai The University of Tokyo
  • Alex Fukunaga The University of Tokyo

DOI:

https://doi.org/10.1609/icaps.v24i1.13620

Keywords:

Abstraction and Reformulation, Robotic Manufacturing, Iterative Plans, Cyclic Scheduling

Abstract

In domains such as factory assembly, it is necessary to assemble many identical instances of a particular product.  While modern planners can generate assembly plans for single instances of a complex product, generating plans to manufacture many instances of a product is beyond the capabilities of standard planners.  We propose ACP, a system which, given a model of a single instance of a product, automatically reformulates and solves the problem as a cyclic planning problem. We show that our domain-independent ACP system can successfully generate cyclic plans for problems which are too large to be solved directly using standard planners.

Downloads

Published

2014-05-10

How to Cite

Asai, M., & Fukunaga, A. (2014). Fully Automated Cyclic Planning for Large-Scale Manufacturing Domains. Proceedings of the International Conference on Automated Planning and Scheduling, 24(1), 20-28. https://doi.org/10.1609/icaps.v24i1.13620