Application of an improved Gaussian-like sum particle filter to large misalignment transfer alignment

On the basis of small misalignment transfer alignment models, a velocity plus attitude matching large misalignment transfer alignment model is treated in detail. For the non-linear and non-Gaussian features of the model, an improved particle filer using Gaussian-like sum resample function is proposed by analyzing the time-domain characteristics of the second-order Markov process. Theoretical analysis shows the unity of large misalignment transfer alignment model and small misalignment transfer alignment model. The simulation with small misalignment verifies the validity of the large misalignment transfer alignment model. And the simulation with large misalignment shows that the accuracy of transfer alignment with non-linear and non-Gaussian features is improved by about 40% with small amounts of increment of calculation.