声速不确定条件下的运动水下航行器自定位

The traditional localization of vehicle using an underwater acoustic positioning system has two problems. One is the motion of the vehicle during observation period is ignored, and the other is that the uncertainty of underwater sound speed is not taken into account. A time measurement model for localization is constructed for the vehicle in motion, and the uncertainty of sound velocity is modeled to solve the above problems. A localization method for the moving vehicle is derived based on weighted least squares estimator, and a sound speed update method based on maximum likelihood estimator is proposed as well. Apart from estimating the vehicle position, the proposed model and the time mesurement are able to update the sound speed. The simulation results verify that the proposed methods outperform the existing methods for various parameter settings and achieve Cramer-Rao lower bound (CRLB) when the time measurement noise is not large. Moreover, the performance of the sound speed update is significantly improved when the time error is not large.