The understanding of computer algorithms can be enhanced through the use of visualisation. This thesis presents the development of a simulator for the evolution of entities in a virtual world. A specific entity is represented by a car that must overcome specific obstacles. The simulator demonstrates the behaviour of individual entities and includes options for setting the parameters of the genetic algorithm, which influence the formation of the next generation. This allows the user to observe how specific parameters affect the evolution of the population. The main challenge of the simulator is to design a car that achieves the longest possible distance. The components of the car are represented as genes, which adapt over generations in order to achieve optimal configurations.
|
