Planejamento online de caminhos por cobertura através de meta-heurística

Abstract

Different practical applications for Unmanned Aerial Vehicles (UAVs) have emerged in recent years, requiring periodic and detailed inspections to verify possible structural changes. Inspections using UAVs should minimize flight time due to battery time constraints and identify topographical features of terrain or structures. In this sense, coverage path planning (CPP) aims to find the best path to cover a given area while respecting the restrictions of the operation. Photometric terrain information is used to create routes or even refine already created paths, in addition to enabling different types of image analysis. Therefore, the main contribution of this research is the development of a methodology that uses a meta-heuristic algorithm to create optimized missions that seek to balance two conflicting objectives: mission time and image quality at 3D reconstructions. The technique was applied both in a simulated scenario and in a real environment to verify its effectiveness, seeking the application of several meta-heuristic techniques and a statistical analysis of the results found. In addition, the algorithm was applied to the most diverse structures, both on slopes and in regions to be investigated, through active sensors such as lasers and maps provided offline through point cloud and digital elevation models. The results showed that the algorithm was able to create optimized missions, equidistant from the surface and with all CPP criteria met. In addition, there is the possibility of controlling the two proposed objectives, aiming to increase the quality of the three-dimensional reconstructions and the mission time.