Master presentation: Heegyeong Kim

Supervisor: Ola Benderius

Examiner: Mats Granath

Searching Pattern Optimization Using A Fleet of Drones

Abstract

The purpose of this thesis is to find the most efficient search pattern for Search and Rescue (SAR) operations at sea using fixed-wing drones following the boat within a communicable distance. Three different search patterns were employed: Lyapunov Guidance Vector Field (LGVF), Bearing Only, and Sinusoidal motion. Genetic Algorithms (GA) were used to find the optimal parameters for each model, and simulations were performed to compare and analyze their effectiveness. The simulations encompassed three distinct scenarios based on the boat's motion: straight, circular, and curved. The results of the simulations were evaluated in terms of the searched area over the required thrust for a single operation and were also used to investigate how the efficiency varied with the number of drones involved. The findings of this study contribute to the field by providing insights into the most effective search patterns for SAR operations using fixed-wing drones, considering different boat motion scenarios.