In-situ Bias Estimation of Low Grade Gyroscopes for Ship Launched Flight Vehicles

Abstract

Abstract The paper presents mechanization of an analytical Transfer Alignment (TA) process and a novel scheme on in-situ bias estimation of Low Grade MEMS Gyroscopes based Inertial Navigation System. The objective of the work presented, is to devise a scheme for estimation of initial vector(s) i.e. velocity, attitude and biases of MEMS gyroscopes of inertial navigation system of flight vehicles being launched from moving ships. Inertial Navigation is a dead reckoning system and prior initialization of velocity and attitude is an essential requirement to achieve minimum Circular-Error-Probable (CEP) at the target point. Establishing the initial vector(s) is straight forward for the flight vehicles being launched from static launchers where the Gravity Vector and Earth Rate Vectors are being measured precisely and utilized for analytical Gyro-Compassing (GC) alignment. The same is not true for moving vehicles like ships, where the reference vectors like Earth Rate and Gravity are not known. Hence, an analytical Transfer Alignment (TA) scheme is developed for pre-initialization of Slave Inertial Navigation System (SINS) on-board flight vehicle with Master Inertial Navigation System (MINS) on-board moving ship. A 5–state Extended Kalman Filter based velocity matching technique is designed and an observability study is presented. The MEMS gyroscopes which have gained importance over their mechanical and optical counterparts are highly reliable, miniature in size and available at low cost. On contrary, these sensors have a poorer day-to-day bias repeatability of the order of 20°/hour which will result into degraded navigation performance. Hence, an attempt was made to devise a scheme to estimate the large biases of SINS, in-situ with external aid from MINS. The presented scheme estimates these gyro biases in tandem with the Transfer Alignment to an accuracy of 1.0°/hour, ensuring convergence within 3 minutes, without compromising on alignment accuracy. A suitable system configuration is conceived based on low grade MEMS gyroscopes and Quartz accelerometers. The total scheme is implemented in real time and experimental results are presented.