Design and Implementation of a Computerized Game (A Case Study Ludo Game)
AIM AND OBJECTIVES OF THE STUDY
The main aim of this study is to design and implement a computerized game which case study is a ludo game.
The objectives of the proposed design include:
- To help in reasoning faster and better.
- To bring fun into people’s homes and offices.
- To act as a good form of relaxation.
- To be able to create a real life feature.
- Relate computer games in various spheres of life e.g. business, art, engineering, science etc.
Today Games and intelligence go side by side. Ludo is an intelligently played mind game. It requires some basic knowledge for different moves that need to be taken on one’s move.
Game theory is the theory that bridges the gap between the complexity and simplicity of the game. A modeling tool has certain set of rules to go ahead with the implementation of game. There are different types of game theories that have been implemented to bring out a solution that is possibly a game from live users to firm users.
Ludo has its base underlying the intelligence of NPC in which the player uses basic Ludo rules to bring that game from human to non-living but machine/programmed players and as intelligently as humans do. (Gomes, n.d)
As game and intelligence have strong connection and brainstorming. Many people have made great effort to build the game into systems.
Georgies (2012) gave a thorough description of Game Theory and its applications. He explained about AI and its four flagships, briefing about the terminology used currently in games and also gave connection between search algorithms with hard combinatorial problems. He grouped this connection under three things planning, duality and randomization.
Briand (1999) illustrated that there has been a complete description about how to solve a game using AI and further emphasize that ludo is implemented a TD(λ)based Ludo player and also implemented a Q-learning based Ludo player using reinforcement learning.
Nowadays, the games are not just stayed to static environment but also they have been developed to intelligent dynamic environment where system acts as second player. Most people enjoy these types of games only where there is a brainstorming part represented. This part is AI (Artificial Intelligence), where the programming is done in such a way that system also act as an intelligent creature, creating interest of users in games.
In Games, AI can be implemented using different types of searches. These may be:
- Heuristic Search
- State Space Search
- Minimax Algorithms
In these algorithms, there is a lot of complexity involved and large trees are made using top down or bottom up approach. Now, if searching deep into the game tree is so hard, how are humans able to play games like Chess and Ludo? Instead humans use some high level heuristic analysis, and ground their moves on experience from previous game play or some sort of intuition (Veenus & Kuldeep, 2015).
Ludo means, “I play”. It is being derived from an ancient game of India, that, is known by the name “Pachisi”. It is a board game for two, three, or four players. In this game the players race against their four tokens from initial to end state according to roll of a dice
Played by four players, Ludo is a board divided into four main areas:
Each player occupies one area of a color. In this he is given four tokens of his chosen color that he race around the board to win.
The main purpose of ludo lies in the comparison of several approaches in the area of artificial intelligence. At one side, Ludo is instead a very simple game and is observable but on the other side, it contains a few challenges due to the stochastic and multi agent environment.
Therefore. it offers a good balance between simplicity and complexity and is able to attract a wide audience and not only professionals. In addition, it is very common and well- known around the world. (Veenus & Kuldeep, 2015).
The methodology adopted in this research work is Q learning. It is a reinforcement learning technique used in machine learning. The goal of Q-Learning is to learn a policy, which tells an agent which action to take under which circumstances. It does not require a special model of the environment and can handle problems with stochastic transitions and rewards, without requiring adaptations.
ANALYSIS OF THE EXISTING SYSTEM
The game can be played by 2, 3 or 4 players and they have to take turns in a clockwise order; highest throw of the die starts.
Each throw, the player decides which piece to move. A piece simply moves in a clockwise direction around the track given by the number thrown. If no piece can legally move according to the number thrown, play passes to the next player.
A throw of 6 gives another turn i.e. a player must throw a 6 to move a piece from the starting circle onto the first square on the track. The piece moves 6 squares around the circuit beginning with the appropriately coloured start square (and the player then has another turn).
If a piece lands on a piece of a different colour, the piece jumped upon is returned to its starting circle and if a piece lands upon a piece of the same colour, this forms a block. This block cannot be passed or landed on by any opposing piece.
When a piece has navigated through the board, it proceeds up the home column. A piece can only be moved onto the home triangle by an exact throw. The first person to move all 4 pieces into the home triangle wins
SYSTEM TESTING AND IMPLEMENTATION
System testing and implementation is the last step in software development. It involves a process of putting into action, a formulated plan. Before implementation, plans must have been completed and objectives must be clear.
SUMMARY, CONCLUSION AND RECOMMENDATION
Today wide range of people are engaged in playing games. Games are most common in all the age groups, whether being a kid or a young or it is an old aged person. Gaming is growing at a faster rate day by day in the field of entertainment. Specially, where artificial intelligence is there that means the end user along with live user is also playing. Therefore, artificial intelligence is most important for making the game more interactive and entertaining for the user. As mobile phones are in boom, we can see games are moving from boards and grounds to hand held screens. Ludo is one such example. Ludo is very common game that is implemented using q learning, which is a type of reinforcement learning. It is optimally customized to reduce the complexity.
Here a brief description of the Game theory has being shown. Along with this, we showed some features of game has been shown and a brief of Artificial Intelligence along with games and usage of AI in games. After this, there was explanation about the gaming searches in algorithms used mainly for the Ludo. And after all this, description of Ludo in the form of case study is explained where the algorithm is optimized for using lesser indices so that there is less complexity for guessing the tack number and then calculation as we follow color indexing along with number and assign.
The current Ludo game that is made is much optimized but still there are chance of improvements by finding even better algorithm that will lead to more optimization. Four more safe states can be added as there is in general board ludo.
Also, we can also implement the game to start from initial by not only getting “6” at dice but also with “1”. Currently, here I have only used continuous rolling of dice after getting “6” as a bonus move. We can also restrict that to “two” only. As we get continuous “three” “6”, the turn is altered without any move of token.
- Briand (1999). A unified framework for coupling measurement in objectoriented systems”. Retrieved from http://web.mit.edu/sp.268/www/gamesearch.pdf on July 16, 2018.
- Gaver et al. (2004). Cultural Probes and the Value of Uncertainty. Interactions. Retrieved from http://jrp.icaap.org/content/v1.1/johansson.html on July 16, 2018
- Georgies (2012). Game AI Revisited. Retrieved from http://yannakakis.net/wp/content/uploads/2012/03/gameAI.pdf on July 15, 2018
- Gomes, (n.d) Structure, Duality, and Randomization: Common Themes in AI and OR. Retrieved from http://www.cs.cornell.edu/gomes/aaai00.pdf on July 14, 2018
- Gregor et al, (n.d). Theory of Fun for Game Designers. Retrieved from http://tampub.uta.fi/tup/951-44-5432-4.pdf on July 15, 2018
- Parcheesi, (2012). A Theoretical Analysis of Cooperative Behavior in Multi-Agent Q-learning Ludo. Retrieved from http://en.wikipedia.org/wiki/Parcheesi on July 14, 2018
- TD(λ) and Q-Learning Based Ludo Players (n.d) Retrieved from http://geneura.ugr.es/cig2012/papers/paper79.pdf on July 16, 2018
- Veenus & Kuldeep, (2015). Artificial Intelligence: Game Techniques Ludo – A Case Study. Retrieved from http://www.krishisanskriti.org/acsit.html on July 15, 2018