Comparison of space launch vehicle tracking using different types of multiple models

Since an instantaneous impact point and impact dispersion area of the space launch vehicle (SLV) are determined from estimated state information (i.e., position and velocity) of the SLV, a precise tracking of the SLV plays an important role in the flight safety operation. To contribute this problem, in this paper, we present comparison of two distinct interacting multiple model (IMM) algorithms which incorporate different types of multiple models, to describe dynamic models of SLV. Here, to design multiple motion models of SLV, we separate the flight phase of a SLV into propelled flight mode and coast (non-propelled) flight mode. The first one is represented by the Singer's linear model with ternary uniform mixtures and the other one is depicted by a nonlinear model with a state vector considering drag and thrust of SLV. To compare the performance of both IMM for SLV, we make use of nominal trajectory of Korea space launch vehicle-1 (KSLV-1) which is the first SLV of Republic of Korea and had launched successfully in 2013. The nominal flight trajectory includes propelled and non-propelled flight modes based on variation of thrust. Through Monte Carlo simulations, we demonstrate the tracking performance of both IMM algorithms in an aspect of RMS error.