Drone Route Optimization using Constrained Based Local Search

• Autor(a) principal: Pimentel, Sérgio Bairos
• Data de Publicação: 2018
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10362/59500
• topic: Drone; Local Search; Farming; Optimization; Route; Coverage Path Planning Problem; Domínio/Área Científica::Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática

This dissertation focuses on an optimization problem that consists of finding the best flight plan (i.e. routes) for an Unmanned Aerial Vehicle (UAV), or drone, overflying a farming field that needs to be swept and sprayed with fertilizers or pesticides. This system calculates an optimal route for a crop field, having into consideration that the field needs to be swept and covered entirely. Drones have been around for many years, especially in the military, but only recently they have expanded to other fields, including agriculture, which makes it a particularly interesting field of study. The drones we consider in this thesis are not any regular type of drone for common use, but rather a specially adapted drone for agriculture to facilitate the process of farming. These drones fly close to the ground spraying fertilizers using specialized extension tools. Therefore, the main problem is to search for an efficient route that sweeps the field and saves as much resources as possible. Additionally, the resolution to this problem should not only consist of searching for an ideal route, but also to help the user in planning and selecting a desired field. Some factors need to be taken into account when considering such problems such as legislation, drone’s energy consumption, and environment related variables such as field elevations or man-made infrastructures. This system is built using local search as a basis, which essentially consists in making small changes to a solution, improving it iteratively until some near optimal solution is hopefully found.