Artificial Neural Network based Intelligence for Omnidirectional Autonomous Underwater Robot  

Oleh:

Vicerra, Ryan Rhay P. ; Bandala, Argel A. ; David, Kanny Krizzy A. ; Fernando, Arvin H. ; Dadios, Elmer P.

Jenis: Article from Proceeding
Dalam koleksi: The 14th Asia Pacific Industrial Engineering and Management Systems Conference (APIEMS), 3-6 December 2013 Cebu, Philippines, page 1-13.
Topik: Underwater Robot; Neural Network; Artificial Intelligence
Fulltext: 4016.pdf (551.82KB)

Isi artikel

In this paper, artificial neural network is used to design the model of an omnidirectional underwater robot. Its purpose is to determine the response of the robot and the necessary command or string of commands for each thruster. The robot platform is a 6-thruster omnidirectional underwater robot design which is also presented in this paper. The neural network is composed of 6 input and 6 output parameters. The input parameters are numerical values which represents the final or target destination along the X, Y and Z axis. Each output parameter corresponds to the individual response of each thruster. The neural network model has 2 layers of feed forward network with 18 sigmoid hidden neurons and 6 linear output neurons. The network is trained using Levenberg-Marquardt backpropagation training algorithm using Matlab.

Opini Anda

Klik untuk menuliskan opini Anda tentang koleksi ini!

Kembali