Solução rápida das ambiguidades para linhas de bases longas com o uso combinado dos sistemas GPS e Galileo
Carregando...
Arquivos
Data
2021-03-24
Autores
Silva, Crislaine Menezes da
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
Os GNSS (Global Navigation Satellite Systems) têm sido amplamente utilizados em atividades de levantamento, geodésia e navegação, bem como em outros campos. Em 2021, há previsão que quatro sistemas globais estejam operacionais, são eles, o americano GPS, o russo GLONASS, o chinês BeiDou e o da União Europeia Galileo. O uso de multi-GNSS pode beneficiar usuários de posicionamento global, navegação e tempo, não apenas em termos de disponibilidade de satélites,mas também precisão, integridade e confiabilidade. Entretanto, para alcançar precisão centimétrica é necessário solucionar corretamente as ambiguidades inteiras nas medidas de fase da onda portadora. A solução das ambiguidades inteiras e a sua validação têm sido questões importantes para posicionamento preciso GNSS ao longo das últimas décadas. Esta tese visa elaborar, analisar e resumir sistemicamente a modelagem e solução das ambiguidades para o posicionamento relativo multi-GNSS de linhas de bases longas, aperfeiçoando e projetando algoritmos, para desenvolver umsistema computacional eficiente, considerando as condições atmosféricas brasileiras. Dentre as contribuições deste trabalho, cabe ressaltar, uma revisão teórica abrangente sobre os algoritmos de solução e validação das ambiguidades e o desenvolvimento de um sistema computacional que realiza o posicionamento relativo de linhas de bases longas, para observaçõesmultifrequência, utilizando os sistemas GPS e Galileo, de forma autônoma ou combinada, que realiza a solução das ambiguidades, considerando a modelagem dos efeitos atmosféricos no Brasil. O desempenho dos modelos e do sistema computacional desenvolvidos é avaliado com dados GNSS reais, em períodos de diferente atividade ionosférica. No primeiro experimento, analisou-se o desempenho do posicionamento relativo utilizando observações dos sistemas GPS, Galileo de forma isolada e combinada. Umconjunto de 4 linhas de bases, de 280ma 144 km, localizadas no estado de São Paulo, num período de junho e outubro de 2019 foi selecionado. Os resultados mostraram que o uso do modelo com ionosfera ponderada teve uma melhoria de 62% na média para as linhas de base longas. Emrelação ao uso combinado dos sistemas GPS e Galileo, a melhoria em comparação ao sistema GPS foi de 24% e em comparação ao sistema Galileo foi de 21%. Visando a solução rápida das ambiguidades, a combinação de duas estratégias foi proposta no segundo experimento: a solução parcial das ambiguidades para o posicionamento relativo com o uso combinado dos sistemas GPS e Galileo. Os resultados foram analisados em relação processamento conjunto dos sistemas GPS e Galileo, juntamente com solução parcial das ambiguidades (PAR). Assim, no segundo experimento foram analisadas duas técnicas de solução das ambiguidades, a solução completa do vetor das ambiguidades (FAR) e a solução parcial (PAR), utilizando dados do sistema GPS e GPS combinado com Galileo, para duas linhas de base, uma curta e uma longa. Os resultados são apresentados em termos de número de satélites rastreados, acurácia do posicionamento, percentual de fixação das ambiguidades e número de ambiguidades fixas. Verificou-se que, no geral, o uso da solução parcial das ambiguidades apresentou um melhor desempenho emrelação a solução completa do vetor de ambiguidades. Para a linha de base curta, foi possível obter a solução instantânea das ambiguidades em todo o período processado utilizando os sistemas GPS e Galileo juntamente coma técnica PAR. Para as técnicas PAR e FAR, o percentual de fixação das ambiguidades foi de 98% e 92% para o sistema GPS e 100% e 95% para o uso combinado dos sistemas, respectivamente. Os resultados obtidos para a linha de base longa, demonstram que o melhor desempenho foi obtido como uso combinado dos sistemas juntamente coma técnica PAR. O tempo para fixas as ambiguidades (TTFF) do sistema GPS foi de 49 min e 12 min para as técnicas FAR e PAR, respectivamente. Para o uso dos sistemas combinados, o TTFF das ambiguidades foi de 33 min e 4 min para as técnicas FAR e PAR, respectivamente. Os resultadosmostram que a solução parcial das ambiguidades pode possibilitar o posicionamento instantâneo centimétrico para estações de referência próximas e reduzir significantemente o tempo necessário para linhas de bases longas em comparação com a solução completa do vetor de ambiguidades.
GNSS (Global Navigation Satellite Systems) have been widely used in surveying, geodesy, and navigation activities, as well as in other fields. In 2021, the four global systems are expected to be operational, which are the American GPS, the Russian GLONASS, the European Union’s Galileo, and the Chinese BeiDou. The use ofmulti-GNSS can benefit users of global positioning, navigation, and timing not only in terms of satellite availability but also accuracy, integrity, and reliability. However, to achieve centimeter accuracy, resolving integer ambiguities in carrier phase measurements is essential. Only if the number of the integer unknown cycles is correctly resolved, the accuracy of centimeter level positioning becomes achievable. Once the integer ambiguities are resolved, the carrier phase measurements begin to act as precise distance measurements. Integer ambiguities resolution and validation have been important issues for GNSS precise positioning over the past three decades. This thesis aims to systemically elaborate, analyze, and summarize the modeling and ambiguity resolution for long-baseline relative multi-GNSS positioning, then improve and design algorithms and develop an efficient software, considering the Brazilian atmospheric conditions. The main contributions of this thesis are a comprehensive theoretical review on the algorithms of ambiguity resolution, and the development of a software that performs RTK positioning of long baselines, using multifrequency observations of GPS and Galileo systems, with ambiguity resolution, considering the modeling of atmospheric effects in Brazil. The performance of the developed models and software is evaluated with real GNSS data, in periods of different ionospheric activity. In the first experiment, the performance of the RTK positioning was analyzed using the GPS and Galileo systems in an isolated and combined way. A set of 4 baselines, from 280 m to 144 km, in São Paulo state and a period of June and October 2019 were selected. The results showed that the use of the weighted ionosphere model had an improvement of 62% on average for the long baselines. Regarding the combined use of the GPS and Galileo systems, an improvement of 24% was obtained compared to the GPS system and 21% compared to the Galileo system. To obtain fast ambiguity resolution, a combination of two strategies was proposed in the second experiment: the joint processing of the GPS and Galileo systems with the partial ambiguity resolution (PAR). The second experiment analyzed two ambiguity resolution techniques, full ambiguity resolution (FAR) and PAR, using GPS-only data and combined with Galileo, for a short and a long baseline. Results were presented in terms of the number of tracked satellites, positioning accuracy, ambiguity success rate, and the number of fixed ambiguities. It was found that, in general, the use of the PAR presented better performance when compared to the FAR. To the short baseline, it was possible to obtain instantaneous ambiguity resolution throughout the processed period using the GPS and Galileo systems together with the PAR technique. For the PAR and FAR techniques, the percentage of ambiguities fixing was of 98% and 92% for the GPS system and 100% and 95% for the combined use of the systems, respectively. The results obtained for the long baseline demonstrate that the best performance was obtained in the combined use of the systems together with the PAR technique. Time to First Fix (TTFF) of ambiguities considering the GPS system was 49 min and 12 min for the FAR and PAR, respectively. For GPS + Galileo, the TTFF of ambiguities was 33 min and 4 min for the FAR and PAR techniques, respectively. The results show that partial ambiguity resolution can enable instantaneous centimeter level positioning to nearby reference stations, and clearly reduces the time required for long baselines compared to full ambiguity vector resolution.
GNSS (Global Navigation Satellite Systems) have been widely used in surveying, geodesy, and navigation activities, as well as in other fields. In 2021, the four global systems are expected to be operational, which are the American GPS, the Russian GLONASS, the European Union’s Galileo, and the Chinese BeiDou. The use ofmulti-GNSS can benefit users of global positioning, navigation, and timing not only in terms of satellite availability but also accuracy, integrity, and reliability. However, to achieve centimeter accuracy, resolving integer ambiguities in carrier phase measurements is essential. Only if the number of the integer unknown cycles is correctly resolved, the accuracy of centimeter level positioning becomes achievable. Once the integer ambiguities are resolved, the carrier phase measurements begin to act as precise distance measurements. Integer ambiguities resolution and validation have been important issues for GNSS precise positioning over the past three decades. This thesis aims to systemically elaborate, analyze, and summarize the modeling and ambiguity resolution for long-baseline relative multi-GNSS positioning, then improve and design algorithms and develop an efficient software, considering the Brazilian atmospheric conditions. The main contributions of this thesis are a comprehensive theoretical review on the algorithms of ambiguity resolution, and the development of a software that performs RTK positioning of long baselines, using multifrequency observations of GPS and Galileo systems, with ambiguity resolution, considering the modeling of atmospheric effects in Brazil. The performance of the developed models and software is evaluated with real GNSS data, in periods of different ionospheric activity. In the first experiment, the performance of the RTK positioning was analyzed using the GPS and Galileo systems in an isolated and combined way. A set of 4 baselines, from 280 m to 144 km, in São Paulo state and a period of June and October 2019 were selected. The results showed that the use of the weighted ionosphere model had an improvement of 62% on average for the long baselines. Regarding the combined use of the GPS and Galileo systems, an improvement of 24% was obtained compared to the GPS system and 21% compared to the Galileo system. To obtain fast ambiguity resolution, a combination of two strategies was proposed in the second experiment: the joint processing of the GPS and Galileo systems with the partial ambiguity resolution (PAR). The second experiment analyzed two ambiguity resolution techniques, full ambiguity resolution (FAR) and PAR, using GPS-only data and combined with Galileo, for a short and a long baseline. Results were presented in terms of the number of tracked satellites, positioning accuracy, ambiguity success rate, and the number of fixed ambiguities. It was found that, in general, the use of the PAR presented better performance when compared to the FAR. To the short baseline, it was possible to obtain instantaneous ambiguity resolution throughout the processed period using the GPS and Galileo systems together with the PAR technique. For the PAR and FAR techniques, the percentage of ambiguities fixing was of 98% and 92% for the GPS system and 100% and 95% for the combined use of the systems, respectively. The results obtained for the long baseline demonstrate that the best performance was obtained in the combined use of the systems together with the PAR technique. Time to First Fix (TTFF) of ambiguities considering the GPS system was 49 min and 12 min for the FAR and PAR, respectively. For GPS + Galileo, the TTFF of ambiguities was 33 min and 4 min for the FAR and PAR techniques, respectively. The results show that partial ambiguity resolution can enable instantaneous centimeter level positioning to nearby reference stations, and clearly reduces the time required for long baselines compared to full ambiguity vector resolution.
Descrição
Palavras-chave
Solução das ambiguidades, Posicionamento relativo, GPS, Galileo, Solução parcial das ambiguidades, Ambiguity resolution, Real time kinematic positioning, Partial ambiguity resolution