A novel partitioned matrix‐based parameter update method embedded in variational Bayesian for underwater positioning

In order to meet the requirements of high-precision positioning for autonomous underwater vehicles (AUVs) in complex and time-varying marine environments, a novel partitioned matrix-based parameter update method embedded variational Bayesian (PMPU-VB) is proposed deal with accuracy problem caused by inaccurate predicted error covariance measurement noise matrices. By employing (VB) method, accurate matrix can be obtained. Subsequently, PMPU-VB, which employs matrix, used as substitute traditional Gaussian filtering (GF) algorithm, probability density function (PDF) state vector. The vector defined follow distribution, parameters distribution are deduced using method. Finally, position information AUV Therefore, more precise acquired. experiments results illustrate that PMPU-VB has higher estimate accuracy, better stability robustness than other comparison algorithms.