Approaching flexible production with planning methods

Abstract

Flexible production systems are becoming more and more important for the industry. Higher requirements for flexibility in the production process demand further improvements in regards to flexible automation. In this perspective, the RoboCup Logistics League competition was designed to provide a testbed for dynamic production domains. The related challenges range from robust navigation over manipulation to efficient production scheduling. In this paper, we provide an update on how the team GRIPS approches the flexible production challenge using modern planning methods. This includes a new long-term planning and scheduling strategy as well as a hierarchical structure which further develops such plans into Behavior Trees.