Publikation

Optimised bumblebee paths as search strategy for autonomous underwater vehicles

Christoph Tholen, Lars Nolle, Tarek El--Mihoub, Oliver Zielinski

In: Ibrahim A. Hameed, Agus Hasan, Saleh Abdel-Afou Alaliyat (Hrsg.). Communications of the ECMS. European Conference on Modelling and Simulation (ECMS-2022) May 30-June 3 Alesund Norway Seiten 107-113 Communications of the ECMS 36 1 ISBN 978-3-937436-77-7 ECMS 6/2022.

Abstrakt

In this paper, the concept of optimised bumblebee (BB) patterns as a search strategy for autonomous underwater vehicles (AUV) is presented. Here, an AUV is used to detect submarine groundwater discharge (SGD) in coastal areas. The optimisation of the BB paths is achieved utilising k-opt optimisation. In this research, 2-opt, 3-opt and 4-opt is used for the optimisation of the BB paths. It is shown using computer simulations that all three optimisation strategies are able to improve the search capabilities of the BB search strategy. The optimisation of the BB path shortens the length of the path to visit the waypoints generated. The saved energy can be used for exploring the search space in more detail, allowing the visit of waypoint the unoptimized BB was not able to reach. The median saved path length is 33.8 m, 43.5 m and 52.6 m for the 2-opt, 3-opt and 4-opt, respectively. The median error over 1,000 experiments of the not-optimised BB is 76.26, while the median error of the optimised BB are 71.63, 72.02 and 72.23 for the 2-opt, 3-opt and 4-opt, respectively.

Weitere Links

2022_ECMS_003.pdf (pdf, 787 KB )

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence