Navnet: AUV Navigation Through Deep Sequential Learning

Xin Zhang,Guangliang Li,Xiaokai Mu,Bo He,Ying Zhou,Tanji Mang

doi:10.1109/access.2020.2982272

Abstract

Achieving accurate navigation and localization is crucial for Autonomous Underwater Vehicle (AUV). Traditional navigation algorithms, such as Extended Kalman Filter (EKF) and Unscented Kalman Filter (UKF), require the system model and measurement model for state estimation to obtain the AUV position. However, this may introduce modeling errors and state estimation errors which will affect the final precision of AUV navigation system to a certain extent. To avoid these problems, in this paper, we proposed a deep framework - NavNet - by taking AUV navigation as a deep sequential learning problem. Firstly, the proposed NavNet can take raw sensor data at different frequencies as input, which benefits from the sequential learning capability of Recurrent Neural Network (RNN). Secondly, NavNet takes advantage of a simplified attention mechanism and Fully Connected (FC) layers to output AUV displacements per unit time, which accomplishes low-frequency AUV navigation by accumulation of it. More importantly, there is no need for the model building and state estimation with NavNet, which avoids the import of relevant errors. We compare the performance of NavNet to EKF and UKF using collected data by running Sailfish in the sea. Experimental results show that NavNet has an excellent performance in terms of both the navigation accuracy and fault tolerance. In addition, a reliable fusion strategy of NavNet and conventional method is applied to achieve high-frequency AUV navigation. The experimental results show that the proposed architecture can be a reliable supplement to limit the error growth of conventional algorithms.

Highlights

In recent years, Autonomous Underwater Vehicles (AUVs) play a crucial role in a tremendous variety of missions, such as oceanographic surveys, marine data acquisition, submarine rescue, and underwater equipment maintenance [1], [2]
AUV DEEP NAVIGATION FRAMEWORK In order to avoid the possible errors imported by the modeling and estimating processes mentioned above, a deep sequential learning framework based on the deep neural networks — NavNet, is proposed for AUV navigation
Since the Global Positioning System (GPS) is available when AUV is carrying out tasks on the surface of the water, we collected the exact location of AUV with GPS as the ground truth

Summary

INTRODUCTION

Autonomous Underwater Vehicles (AUVs) play a crucial role in a tremendous variety of missions, such as oceanographic surveys, marine data acquisition, submarine rescue, and underwater equipment maintenance [1], [2]. Clark et al [33] put forward a VINet to motion estimation using visual and inertial sensors for Visual Inertial Odometry (VIO) which integrated the data at an intermediate feature-representation level In their work, both the Inertial Measurement Unit (IMU) data and monocular RGB images are taken as input for the proposed CNN-RNN model to estimate the poses. Different from the regular methods that work on dealing with state estimation errors, in this paper, we propose NavNet, a deep framework based on deep neural networks for AUV navigation. Since the raw sensor values are taken as input for the deep framework to get the AUV position without using a specific system model, the inevitable errors introduced by the modeling process can be prevented which results in an improvement on the the accuracy of AUV navigation.

AUV PLATFORM

MODEL BUILDING

AUV DEEP NAVIGATION FRAMEWORK

EXPERIMENTAL RESULTS AND ANALYSIS

CONCLUSION