A Biologically Inspired Neural Network for Solar Powered Autonomous Surface Vehicles

Abstract

This paper describes a neural network model for the reactive behavioural navigation of an autonomous surface vehicle (ASV) in which an innovative, neurobiological inspired sensing control system and a hardware architectures are being implemented. The ASV is used to power and support for a Unmanned Underwater Vehicle (UUV), which incorporates several types of environmental and oceanographic instruments such as CTD sensors, chlorophyll, turbidity, optical dissolved oxygen (YSI V6600 sonde) and nitrate analyzer (SUNA) together with ADCP, side scan sonar and video camera. The ASV gets its energy through solar photovoltaic modules, also has automatic devices for the deployment and collection of underwater robots. Navigation system contains accelerometers, gyroscopes, magnetometers and GPS, to reach an appropriate level of spatial location at all times, and corrects trajectory using a neural control algorithm to process the corresponding corrections.