Improving parameter estimation efficiency of a non linear manoeuvring model of an underwater vehicle based on model basin data

In this work, a methodology is proposed for the improvement of the parameter estimation efficiency of a non-linear manoeuvring model of a torpedo shaped unmanned underwater vehicle. For this purpose, data from different tests, were carried out with the aforementioned vehicle at the facilities of the CEHIPAR (Canal de Experiencias Hidrodinámicas de El Pardo), Madrid. In the proposed methodology, the following aspects are taken into account in order to improve the parameter estimation efficiency: selection of the sampling period, smoothing of the data acquired in the tests, considering a compromise between bias and variance of the smoothing filter to be applied, analysis of the classical linear regression model proposed in each trial from the statistical point of view for the estimation of the parameters. Improvements in efficiency are verified by graphical and statistical methods.