Using The CMA Evolution Strategy For Locating Submarine Groundwater Discharge
For effective localisation of a search target by a swarm of Autonomous Underwater Vehicles (AUVs), a suitable cooperative search strategy should be utilised. Various aspects of the search task should be taken into account when selecting a search strategy. The nature of the search environment, the se...
Gespeichert in:
| Veröffentlicht in: | International ECMS Conference on Modelling and Simulation (34. : 2020 : Wildau, Landkreis Dahme-Spreewald) 34th European Conference on Modelling and Simulation (ECMS 2020) |
|---|---|
| 1. Verfasser: | |
| Weitere Verfasser: | , |
| Pages: | 34 |
| Format: | UnknownFormat |
| Sprache: | eng |
| Veröffentlicht: |
2020
|
| Schlagworte: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Tag hinzufügen | Zusammenfassung: | For effective localisation of a search target by a swarm of Autonomous Underwater Vehicles (AUVs), a suitable cooperative search strategy should be utilised. Various aspects of the search task should be taken into account when selecting a search strategy. The nature of the search environment, the search target and the search agents should be considered. The Covariance Matrix Adaption Evolution Strategy (CMA-ES) is a well-known search strategy that proves its success in solving different continuous optimisation problems. This paper investigates utilising the CMA-ES to locate a Submarine Groundwater Discharge (SGD) using the temperature of water as a tracer. The impact of introducing some of the constraints, which are imposed by the search task, on the CMA-ES performance are studied. The influence of the number of the AUVs and their energy capacities on the search performance is investigated. The effect of the resolution of the temperature sensors together with the localisation and the navigation problems on the search behaviour are explored. The results show that these constrains have varying degrees of impact on the performance of the search strategy. |
|---|---|
| ISBN: | 9781713813682 |