Traditional marble game using ant colony optimization / Muhammad Izzat Imran Che Isa

• Main Author: Che Isa, Muhammad Izzat Imran
• Format: Thesis
• Language: English
• Published: 2017
• Subjects: Instruments and machines; Electronic Computers. Computer Science; Evolutionary programming (Computer science). Genetic algorithms; Computer software; Application program interfaces; Configuration management; Development. UML (Computer science); Software measurement; Algorithms; Database management
• Online Access: https://ir.uitm.edu.my/id/eprint/69623/1/69623.pdf; https://ir.uitm.edu.my/id/eprint/69623/

The traditional marble game is a stochastic game which is has random probability of distribution or pattern that may be analysed statistically but not be predicted precisely. It need to target which marble is the most suitable to shot accurately. The study of traditional marble game will be implemented in a game prototype using Ant Colony Optimization (ACO). ACO technique is used for searching method in order to find the nearest marble that can be selected to be shot. Instead of get random value of position, ACO helps to optimize the nearest marble. Collision detection technique is one of a physic law that manipulate the movement and position of marble. The prototype is evaluated based on distance between initial position and selected marble position. The shortest distance is defined as best result. A set of data contains position of marbles is tested for ACO algorithm in the prototype. As the result, the accuracy of the shortest distance is moderate because number of marbles limited to fifty marbles. The more number of marbles, the higher the search accuracy. In future, traditional marble game can be applied with other search algorithm to optimize the solution