Cricket AI - Artificial Intelligence in a Framework

Samuel Gibbs


 The task that has been undertaken in this project is that of determining what type of artificial intelligence is the best one to be used in a cricket game. The aimed type of cricket simulator is one for the twenty over (Twenty20) format because this is the newest and as a result least well simulated (in the author’s opinion) of all the formats of cricket by commercially available simulators and games.  The project is using Sun Microsystems Java as the programming language of choice. This will be done by testing a variety of artificial intelligence techniques and finding out which is best.  Also, by building the framework itself, working out what features it needs to make it playable and interesting. These will include at least artificial intelligences that can play the game against a human and if possible a way of selecting which players will play the game. There will also be some technical research done into the different artificial intelligence techniques to help decide which methods should be attempted. To make the framework better there will be testing done of other commercial products in the cricket management and simulator field, this will also give a better understanding of what should be included within the project. The simulator should simulate a game if cricket with batting, bowling and fielding all simulated. The game simulated will initially be between England and Australia, although there is no reason why other teams couldn’t be used.



Computing Courses

BSc and HND Computing Provision (click on the links below for more details of the courses)



    The University of Northampton's, Department of Computing and Immersive Technologies offers five courses within the MSc Computing postgraduate provision (shown below) all available either part-time or full-time. 

    The contents are the opinion of the author(s) and not necessarily the view of the University of Northampton.