A neural network model for a 5-thruster unmanned underwater vehicle

• Main Authors: Simbulan, Kristan Bryan, David, Kanny Krizzy, Vicerra, Ryan Rhay P., Atienza, Rumel, Dadios, Elmer P.
• Format: text
• Published: Animo Repository 2012
• Subjects: Autonomous underwater vehicles; Intelligent agents (Computer software); Neural networks (Computer science); Electrical and Computer Engineering; Electrical and Electronics; Systems and Communications
• Online Access: https://animorepository.dlsu.edu.ph/faculty_research/1911
• Institution: De La Salle University

Unmanned underwater vehicles (UUVs) are mostly used for safe underwater explorations and researches. UUVs are subject to different parameters that changes over time. Such parameters are not considered in kinematic modelling of vehicles. As such, a dynamic modelling of underwater vehicles is necessary. This study proposes a dynamic model that is utilizing Artificial Neural Network (ANN), for a 5-thruster underwater vehicle design. The training data for the ANN model is gathered by empirical methods. The dynamic model is represented by UUV variables: thrusters input voltages and resulting velocity vector. The results of the neural network showed accuracy and reliability due to the low Mean Square Error (MSE) and satisfactory regression plots. © 2012 IEEE.