Estimating Dynamic Motion Parameters With an Improved Wavelet Thresholding and Inter-Scale Correlation

Abstract

With the challenging noisy measurements, the maneuvering safety and comfort level of hovercraft navigation can be significantly increased by accurate motion parameter estimation. Because traditional hard/soft thresholding schemes have discontinuities and deviation shortcomings, an improved wavelet thresholding method is proposed to smooth and estimate the dynamic motion parameters. Then, an interscale correlation method is introduced to improve the accuracy of judgment of the wavelet coefficients near the threshold. The newly improved wavelet thresholding and inter-scale correlation scheme can increase the adaptability of the estimator function and improve the accuracy of judgment of wavelet coefficients. Afterwards, a series of numerical simulation experiments are carried out by using four benchmark signals with three different noise levels. Compared with traditional hard/soft thresholding and semi-soft thresholding methods, the proposed improved parameter estimation scheme has a great advantage in denoising the signal noise and reducing the peak interference and the error between decomposition signal and the original signal. Finally, the proposed improved parameter estimation scheme was applied to a certain hovercraft in sea trial. Representatively, some important safety and comfort indicator parameters are selected from actual hovercraft motion measurements to verify the effect of the proposed scheme. From the data analysis, it is shown that the improved scheme behaves an outstanding performance in denoising and estimating for practical application, which improves the safety and comfort of navigation. Otherwise, the effective estimation of indicator parameters will make the further active control be available to assure maneuvering safety and comfort level of hovercraft navigation.

Keywords

• Motion parameters
• wavelet transform
• thresholding function
• inter-scale correlation
• denoise