Localization System for Autonomous Underwater Vehicle during Operations in Lakes of Hydroelectric Dams

Abstract

This article addresses the problem of autonomous underwater vehicle navigation during operations in lakes of hydroelectric dam with respect to the acoustic positioning system. There are numerous challenges to the navigability of these vehicles, particularly the acoustic sensing system, used to locate them and obtain information about environmental conditions. Notably, the environment of hydroelectric's lakes are characterized by a large area with different characteristics regarding the water temperature, large amount of suspended organic matter, vegetation, noise and currents of different intensities and directions. It is proposed to investigate an alternative configuration with the aid of statistical learning methods such as Support Vector Machines for the localization of these vehicles during the operation in this environment. In the future, the use of these vehicles will enhance the characterization of the lake of Binacional ITAIPU, by collecting data relevant to conservation of this important asset.