Procedimentos metodológicos para atualização de informações geoespaciais por meio de sensores imageadores embarcados em sistema aéreo remotamente pilotado

• Ano de defesa: 2022
• Autor(a) principal: Ramalho, Breno da Silva
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Tecnológica Federal do Paraná; Curitiba; Brasil; Programa de Pós-Graduação em Engenharia Civil; UTFPR
• Palavras-chave em Português: Drone; Dados geoespaciais; Fotogrametria aérea; Cartografia - Processamento de dados; Sensoriamento remoto; Sistemas de informação geográfica; Drone aircraft; Geospatial data; Aerial photogrammetry; Cartography - Data processing; Remote sensing; Geographic information systems; CNPQ::ENGENHARIAS::ENGENHARIA CIVIL::CONSTRUCAO CIVIL; Engenharia Civil
• Link de acesso: http://repositorio.utfpr.edu.br/jspui/handle/1/30308

Unmanned aerial vehicles (UAV), also known as Remotely Piloted Aircraft (RPA or ARP), or popularly called drones, have been gaining ground in the world economy. Its use through aerial photogrammetry has particularly grown in the construction industry, to help corporations improve accuracy and productivity on the job site. Aerophotogrammetry with ARP is a low-cost technique for producing maps with high spatial and temporal resolution. However, the diversity of equipment, sensors, data collection and processing methodologies requires a systematic analysis in order to ensure that the products can be used to generate and update heritage digital cartographic information, demanded by the Brazilian Army (EB). Seeking technological development, EB has produced management tools with an emphasis on information technology. Thus, the Unified System of the Works Process (OPUS) was developed in the military sphere, which aggregates all the information regarding the execution of civil engineering works in military installations. Such information requires continuous updating and there is no methodological procedure that contemplates such action in that institution. This study analyzed the possibility of performing the geospatial update, using information from digital cartography registered in the OPUS system in accordance with national cartographic legislation. It was also verified if there was a loss of planimetric precision of the updated data, as well as if the photogrammetric products (orthoimages), coming from image sensors embedded in a remotely piloted aerial system (RPAS) meet the parameters contained in the Cartographic Accuracy Standard of Cartographic Products Digital – PEC-PCD (Decree No. 89,817/84). The study has two areas of action that were called experiments 1 and 2, carried out in Military Organizations (OM) located in the municipalities of Lapa and Curitiba, both in the State of Paraná, with approximate areas of 0.19 and 0.21 km2. , with 175 and 84 improvements registered in the OPUS system, respectively. The geospatial update was performed using the free software QGIS. In order to achieve the results, planimetric discrepancies were calculated to enable the detection of outliers, trend analysis (spatial distribution, normality and trend tests) and spatial precision calculated using the GeoPEC software. The calculated planimetric discrepancies for experiments 1 and 2 presented the following values for their positional accuracies, respectively: mean of 5.8 cm and 4.3 cm, standard deviation of 2.8 cm and 1.9 cm, RMS of 6 .4 cm and 4.7 cm. As results of the study, point clouds, orthophotocharts of each area, digital surface (MDS) and terrain (MDT) models, contour lines with metric equidistance and orthoimages for each experiment were generated. 100% of the information registered in the shapefiles were also updated. The positional accuracy of the generated orthoimages was evaluated, being considered positionally accurate, classified as Class A according to the PEC-PCD, with their satisfactory planimetric accuracies (Mean Squared Error – RMS), meeting the two conditions of Decree-Law no. 89,817/ETADGV, unbiased and spatially accurate data, to the scale of 1:1,000, for both experiments.