Automated Construction of Transferable Loading Algorithm with Cartesian Genetic Programming
Abstract
This paper proposes an automatic construction technique of transferable loading algorithms based on Cartesian Genetic Programming. The proposed method aims to construct the loading algorithm with a few hundred fitness evaluations by optimizing the execution order of rules to decide a type of multiple cardboard boxes and their loadable positions simultaneously. Experimental results show that auto-constructed loading algorithms can derive competitive performances to defined baselines under two hundred fitness evaluations on similar problems without any additional fitness evaluation.
Publication
Transactions on Mathematical Modeling and its Applications, Vol. 14, No. 3, pp. 11–26 (in Japanese)
Notice
- All materials on this page are author’s versions, not necessarily coincide with final published versions.
- IPSJ issued the call for paper of this journal in conjunction with the 131st MPS, a national conference.
Kei Nishihara
3rd-year Doctoral Student
My research interests include evolutionary computation and swarm intelligence.