Coarse Alignment of a Ship's Strapdown Inertial Attitude Reference System Using Velocity Loci

This paper presents a novel estimation method for fast initial coarse alignment of a ship's strapdown inertial attitude reference system using only inertial measurement unit (IMU) measurements for quasi-static alignment and IMU measurements with GPS aiding for moving-base alignment. Unlike several current techniques, the presented estimation method is effective with any initial attitude error. The estimator is based on the decomposition of the attitude quaternion into separate Earth motion, inertial rate, and alignment quaternions. The alignment quaternion is estimated using a minimum variance fit between loci of body- and navigation-frame velocity vectors using solutions to Wahba's problem. One set of vectors is derived from time integrals of measured vehicle motions, and the second set is derived from Earth motion and GPS data (when moving). For the case of quasi-static alignment, an algebraic expression for the covariance of the attitude estimate as a function of the variance of navigation-frame velocity disturbances is developed. It is shown that, by averaging and interleaving the velocity vectors, the resulting attitude estimate is improved over sequential sampling techniques. It is further shown, for a maneuvering vessel, that a continuous estimate of the attitude error covariance can be generated from the IMU data. This latter feature allows direct initialization of a follow-on fine-alignment stochastic estimator's covariance matrix. Results are presented for quasi-static alignment using inertial sensors only and for full in-motion alignment using navigation-frame GPS velocity and position aiding.