No abstract

This study aims to address the rapid transfer alignment (RTA) issue of an inertial navigation system with large misalignment angles. The strong nonlinearity and high dimensionality of the system model pose a significant challenge to the estimation of the misalignment angles. In this paper, a 15-dimensional nonlinear model for RTA has been exploited, and it is shown that the functions for the model description exhibit a conditionally linear substructure. Then, a modified stochastic integration filter (SIF) called marginal SIF (MSIF) is developed to incorporate into the nonlinear model, where the number of sample points is significantly reduced but the estimation accuracy of SIF is retained. Comparisons between the MSIF-based RTA and the previously well-known methodologies are carried out through numerical simulations and a van test. The results demonstrate that the newly proposed method has an obvious accuracy advantage over the extended Kalman filter, the unscented Kalman filter and the marginal unscented Kalman filter. Further, the MSIF achieves a comparable performance to SIF, but with a significantly lower computation load.

The performance of an inertial navigation system (INS) operated on a moving base greatly depends on the accuracy of rapid transfer alignment (RTA). However, in practice, the coexistence of large initial attitude errors and uncertain observation noise statistics poses a great challenge for the estimation accuracy of misalignment angles. This study aims to develop a novel robust nonlinear filter, namely the stochastic integration H∞ filter (SIH∞F) for improving both the accuracy and robustness of RTA. In this new nonlinear H∞ filter, the stochastic spherical-radial integration rule is incorporated with the framework of the derivative-free H∞ filter for the first time, and the resulting SIH∞F simultaneously attenuates the negative effect in estimations caused by significant nonlinearity and large uncertainty. Comparisons between the SIH∞F and previously well-known methodologies are carried out by means of numerical simulation and a van test. The results demonstrate that the newly-proposed method outperforms the cubature H∞ filter. Moreover, the SIH∞F inherits the benefit of the traditional stochastic integration filter, but with more robustness in the presence of uncertainty.

This study aims to address the accuracy evaluation problem for rapid transfer alignment with the coexistence of large misalignment angles and uncertain observation noises. For the requirement of accuracy evaluation, complete information in terms of misalignment angles should be estimated during the alignment process. Thus, a fixed-interval smoothing approach is the core of solving this problem. In this paper, a new Divided Difference Filter (DDF)-based an Interacting Multiple Model Two-Filter Smoother (IMM-TFS) is developed to estimate the misalignment angles. The proposed DDF-based IMM-TFS releases the restriction of inverse nonlinearity by using the weighted statistical linearization regression method, and the resulting pseudo-linear model can be used for backward-time IMM filtering. The smoothing step takes into account the merging of estimations and the interaction of multiple models simultaneously. The new smoother is compared with the previous well-known methodologies in simulations. The results show that the DDF-based IMM-TFS can achieve better accuracy for misalignment angles estimation, and has a high efficiency for detecting the changes in a model.

Cantonese clock manufacture technique is the precious cultural wealth of the Chinese nation. It is necessary to take measures for better protection and inheritance of traditional culture such as Cantonese clock and to present the traditional intangible cultural heritage in a form that meets the needs of modern life and is combined with the development of the times and modern life style, which is of far-reaching significance for building a strong economic and cultural country.