Feed forward gradient descent with momentum backpropagation neural network in tower defense targeting system

Ever since the birth of Adobe Flash, tower defense games became more and more popular. In a tower defense game, you'll need to protect your base from incoming waves of enemies by building towers that will attack the enemies whenever it is in its range. There are different ways the tower can sel...

Full description

Saved in:
Bibliographic Details
Main Author: Ramos, John Ernest D.
Format: text
Language:English
Published: Animo Repository 2013
Subjects:
Online Access:https://animorepository.dlsu.edu.ph/etd_bachelors/2633
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: De La Salle University
Language: English
Description
Summary:Ever since the birth of Adobe Flash, tower defense games became more and more popular. In a tower defense game, you'll need to protect your base from incoming waves of enemies by building towers that will attack the enemies whenever it is in its range. There are different ways the tower can select which enemy it will attack, but none of which coordinates with each tower. In this thesis a new targeting system that uses an artificial neural network to select an enemy and coordinate with other towers was proposed. By comparing it to different implementations of a tower's targeting system, it was proved that an artificial neural network may be used in order to select a target creep and coordinate with other built towers, that results to a fewer life lost, higher killed enemies and more gold earned.