On the estimation of roll alignment error of an ultra short baseline navigation system

This study proposes an algorithm to estimate the alignment errors of an ultra short baseline (USBL) navigation system. This algorithm is based on positioning errors caused by heading, pitch, and roll misalignments, respectively, when running a straight-line USBL transponder positioning. The positioning errors arising from each of the angular misalignments were derived. The differences of positioning error arising from each of the angular misalignments were identified and, accordingly, these differences outlined a simple and intuitive iterative scheme to calibrate each misalignment angle in turn. In addition, the effect of measurement error on the estimation of roll misalignments was investigated. The criterion to obtain a robust estimation of the roll alignment error was provided. Two practical problems associated with vessel motion along a straight-line path were considered. These two problems encountered in field operations refer to course deviation and cross-track error, and they were analyzed and solved to improve the accuracy of estimation. As a result, a vessel without the capability of dynamic positioning (DP) can be run to collect USBL observations for misalignment calibration. A field experiment was conducted and the experimental results demonstrate the validity and high efficiency of the proposed estimation approach. The experimental results also show that an inaccurate estimation of roll alignment error will significantly degrade the quality of estimations on heading and pitch alignment errors.