A Multiple Model Approach for Estimating Roll Rate of a Very Fast Spinning Artillery Rocket

Adding guidance to artillery rocket and munition is important to improve CEP (circular error probability) of unguided weapons. However, high spinning rate is a prohibitive obstacle to use common guidance methods for navigation and attitude determination for unguided rockets and munitions. Finding efficient but cheap guidance solutions for these systems is an emerging topic. In order to investigate possible navigation solutions, rotational rate of the rocket in a mission flight has to be analyzed. In this work, a 2-axis magnetometer and a GNSS receiver system is utilized with interacting multiple Kalman filters to estimate the roll rate of a 122 mm artillery rocket. A flight test is carried out and the real data of the flight test is used to create base and simulated data for Monte Carlo simulations. It is shown that with the multiple model approach presented in this paper, the mean roll rate errors due to measurements decreases from 50 °/s to 15 °/s during the flight of the rocket.