Análise da escalabilidade de método bioinspirado de exploração e mapeamento de ambientes internos desconhecidos usando times de VANTS

Abstract

This work analyzes the scalability of a bio-inspired method for exploring and mapping unknown indoor environments using multiple Unmanned Aerial Vehicles (UAVs) that act in a coordinated manner to complete an exploration mission in the shortest possible time. To achieve this, this work uses NetLogo as a simulation tool, which enables the creation of scenarios and the implementation of the method proposed by (ROSA, 2020). This method uses a topological structure to represent the map inspired by the hexagonal shape of bee hives, só that each hexagon is a new space to be mapped and explored. The scalability analysis uses three different strategies for allocating hexagons to be explored: First In First Out (FIFO), Euclidean Distance (DE) and Relative Euclidean Distance (DER). The proposed scalability consists of adding multiple UAVs to the exploration team, following the Fibonacci sequence, reaching up to eighty-nine UAVs. The work by (ROSA, 2020) has used only three UAVs. As in the original work, the Relative Euclidean Distance (DER) strategy presented more satisfactory results in terms of traffic intensity and total time spent exploring and mapping the environment. However, it can be seen that in scenarios that correspond to the same size as the original, the performance improvement with the increase in the number of UAVs occurred up to a certain maximum number of UAVs in the exploration team, depending on the strategy used. Above a certain amount, the performance improvement is not significant (or even worsens) due to space limitations.