Study on GPS attitude determination technology based on QPSO algorithm

Xia, Kewen; Zhang, Xinying; Gao, Jinyong; Zhang, Ling

doi:10.1109/wcica.2008.4593208

Cited by 4 publications

(1 citation statement)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The methods of solving integer ambiguity include the ambiguity function method [10], the least-squares search method [11], the fast ambiguity solution method [12] and the leastsquares ambiguity decorrelation adjustment (LAMBDA) method. Among all methods, the LAMBDA method is widely recognized as the most rigorous in theory and the most efficient to search for the optimal solution [13,14].…”

Section: Introductionmentioning

confidence: 99%

Parameterization-switching GNSS attitude determination considering the success rate of ambiguity resolution

Zhang

Chang

Chen

et al. 2020

Meas. Sci. Technol.

View full text Add to dashboard Cite

The two key elements of GNSS attitude determination are the correct integer ambiguity resolution (IAR) and the choice of appropriate method. The least-squares ambiguity decorrelation adjustment method is normally used to fix ambiguities using float estimates. However, it is difficult to get all ambiguities fixed in real time. Therefore, a method of partial ambiguity resolution which considers the success rate of IAR and makes full use of the ambiguity information at each epoch is proposed. The ambiguities after decorrelation can be divided into fixed ambiguities and unfixed ambiguities considering the fixing success rate. At the next epoch, the fixed ones are treated as constants, while the float estimates of the unfixed ones, along with the corresponding covariance matrix, are treated as prior information or pseudo-measurements. The parameters of the attitude determination problem can be baseline coordinates (BCs) or attitude angles (AAs). BC parameterization, without considering the baseline constraints, has the merit of resulting in a linear model; however, it involves parameter redundancy, which reduces the model strength. AA parameterization, although avoiding parameter redundancy, brings in nonlinearity and hence linearization errors in the least-squares solution. The linearization errors decrease as the number of fixed ambiguities increases. A switching-parameterization method is proposed; namely, at any epoch, as long as there are at least three fixed double-difference ambiguities, AA parameterization is adopted, otherwise BC is adopted. Even with AA, BC is firstly estimated and then transformed to attitude estimates at which the nonlinear measurement equations are linearized and solved. To verify the performance of the proposed method, a comparative study is implemented with BC and AA methods in the static experiment. The results show that the proposed method can fix ambiguity faster and have better stability and accuracy. In addition, the yaw angle is consistent with the actual running route of the vehicle in the kinematic experiment.

show abstract

Section: Introductionmentioning

confidence: 99%