Deterministic Planning for Flexible Intralogistics

Maximilian Berndt; Christoph Fischer; Dennis Krummacker; Hans Dieter Schotten

In: The 26th IEEE International Conference on Automation and Computing 2021. IEEE International Conference on Automation and Computing (ICAC-2021), System Intelligence through Automation and Computing, September 2-4, Portsmouth, United Kingdom, IEEE, 2021.


An automated planning unit that enables the user to deterministically schedule transportation tasks for intralogistics use cases is proposed. The developed solution aims at inducing a high degree of determinism into transportation task planning in manufacturing industries while at the same time providing the user with the opportunity to flexibly react to rapidly changing constraints, such as updated order situations. The main objective of the software tool is to facilitate the order management process and ensure conflict-free path planning and following of a centrally guided fleet of mobile robots serving transportation tasks. Furthermore, in order to meet customer demands in terms of responsiveness to altered circumstances, the system is able to re-allocate already planned transportation tasks in favor of more urgent ones that may come in without further notice. This is achieved by adopting concepts commonly used in real-time operating systems to the complex problem of intralogistics task scheduling. Sporadically incoming transportation tasks are scheduled dynamically with regard to deadlines, priority levels, available resources as well as estimated execution effort. Flexibility and system responsiveness are increased noticeable by applying on-line task migration mechanisms. The eligibility of the adapted concepts is demonstrated by deploying the proposed solution to test cases within a simulation environment. For this purpose a scalable \gls{kpi} function is proposed as well.


Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence