Performance Improvement of Laser TRN using Particle Filter Incorporating Gaussian Mixture Noise Model and Novel Fault Detection Algorithm
Kyung Jun Han, Chang-Ky Sung, Myeong Jong Yu, Chan Gook Park · Proceedings of the Satellite Division's International Technical Meeting (Online)/Proceedings of the Satellite Division's International Technical Meeting (CD-ROM) · 2018
Terrain referenced navigation (TRN) has been actively researched as an alternative to the GNSS denied environment. The operation principle of the TRN algorithm is to estimate the current position by comparing the measured terrain heights with the digital elevation model (DEM). The problem of radar altimeter is the large footprint that induces large measurement errors. This problem can be eliminated by replacing the radar altimeter with a laser altimeter. Laser TRN expects more accurate performance than radar based TRN due to its small footprint. However, laser measurements are vulnerable to defects according to environmental condition. In this paper, we propose two methods to improve the performance of laser TRN. The first is to accurately model the laser measurement error using the Gaussian mixture noise model. The statistical characteristics of the laser measurement error are considered. Secondly we suggest an effective fault detection algorithm to eliminate the outliers in measured altitude. Large, irregular measurement errors severely degrade TRN performance and impair system stability. Proper rejection of outliers is a crucial factor for improving TRN performance. The fault detection algorithm consisting of the residual sequence check and stochastic terrain roughness check is formulated in the particle filter framework. Through simulation experiment we show that the proposed fault detection algorithm effectively eliminates singularities and improves the performance and reliability of TRN. The proposed fault rejection scheme is applicable to any kind of nonlinear Bayesian state estimators.