Underwater inertial navigation with long base line transceivers: A Near-Real-Time approach

Abstract

Due to the short range over which electromagnetic fields decay in sea water, inertial navigation systems for underwater vehicles are often aided by acoustic time-of-flight positioning scheme. One widely implemented long-baseline (LBL) approach uses a ping-response protocol resulting in asynchronous measurements that depend on the state of the vehicle at two time instants. Due to these issues, the standard assumptions necessary for Extended Kalman Filter (EKF) solutions are not satisfied. This paper proposes a Near-Real-Time (NRT) Bayesian smoothing framework for the LBL aided INS application. Within this NRT framework, the navigation process is divided into the ping-response cycles of LBL transceiving. Before the end of a LBL cycle, a traditional realtime EKF is implemented using the IMU and standard aiding measurements. At the end of each LBL cycle, an optimal Bayesian trajectory estimator executes. This Maximum-A-Posteriori (MAP) estimation includes all the measurement information collected during the current LBL cycle. Furthermore, right after this smoothing process, the current EKF estimate is corrected by the corresponding MAP estimate. This article presents the theoretical solution, discusses the implementation, and presents simulation results to illustrate the accuracy and reliability of this Near-Real-Time approach.