An Underwater Integrated Navigation Algorithm to Deal With DVL Malfunctions Based on Deep Learning

Ding Li,Jiangning Xu,Miao Wu,Hongyang He

doi:10.1109/access.2021.3083493

Ding Li, Jiangning Xu + Show 2 more

Open Access

https://doi.org/10.1109/access.2021.3083493

Copy DOI

Journal: IEEE Access	Publication Date: Jan 1, 2021
Citations: 24	License type: CC BY 4.0

Affiliation: Naval University of Engineering

Abstract

In underwater navigation systems, Global Navigation Satellite System (GNSS) information cannot be used for navigation. The mainstream method of autonomous underwater vehicles (AUV) underwater navigation system is Doppler Velocity Log (DVL) aided strapdown inertial navigation system (SINS). However, because the DVL is an instrument based on Doppler frequency shift to measure velocity, it is easily affected by the external environment. In a complex underwater environment, DVL output is easily polluted by outliers or even interrupted. In this paper, A new integrated navigation algorithm based on deep learning model is proposed to deal with DVL malfunctions. First, use RKF based on Mahalanobis distance algorithm to eliminate outliers, and then train the Nonlinear AutoRegressive with eXogenous input (NARX) model when DVL is available. When DVL is interrupted, use the NARX model to predict the output of DVL and continue integrated navigation. The proposed NARX-RKF scheme’s effectiveness verification was performed on the data set collected by the SINS/DVL ship-mounted experimental system. For comparison, different methods are also compared in the experiment. Experimental results show that NARX-RKF can effectively predict the output of DVL and is significantly better than other methods.

Highlights

High-precision, high-reliability underwater navigation and positioning technology have vital in autonomous underwater vehicles (AUV)
This paper focuses on a Strapdown Inertial Navigation System (SINS)/Doppler Velocity Log (DVL) integrated navigation algorithm based on Nonlinear AutoRegressive with eXogenous input (NARX) and RKF
Unlike the open-loop integrated navigation, the velocity measured by DVL and the velocity calculated by SINS is processed by Kalman filtering in real-time and compensated to SINS in real-time so that the attitude, velocity and position output by SINS are both passed the best estimate result

Summary

INTRODUCTION

High-precision, high-reliability underwater navigation and positioning technology have vital in autonomous underwater vehicles (AUV). Reference [14] attempts to use ANN to completely replace geometric algorithms, that is, to use ANN-based inertial measurement algorithms to directly generate motion states from IMU data It can fundamentally eliminate errors caused by integration, it loses the important advantages of inertial navigation: short-term accuracy and reliability. Li et al [22] used the Nonlinear AutoRegressive with eXogenous input (NARX) model to predict the DVL failure navigation output during the SINS/DVL integrated navigation, but this network only selected the velocity increment as the network input, which did not allow the NARX model to learn the SINS error law well, so the long-term error still increases rapidly when DVL is interrupted. The experimental results show that the method proposed in this paper can effectively suppress the pollution of outliers and effectively perform integrated navigation when DVL fails.

AUV INTEGRATED NAVIGATION SYSTEM MODEL

ERROR MODEL OF SINS

ERROR MODEL OF DVL

ROBUST KALMAN FILTER

Findings

NARX MODEL OF INTEGRATED NAVIGATION SYSTEM