Optimising Agent Behaviours and Game Parameters to Meet Designer’s Objectives

The game industry is one of the biggest economic sector in the entertainment business whose product rely heavily on the quality of the interactivity to stay relevant. Non-Player Character (NPC) is the main mechanic used for this purpose and it has to be optimised for its designated behaviour. The development process iteratively circulates the results among game designers, game AI developers, and game testers. Automatic optimisation of NPCs to designer’s objective will increase the speed of each iteration, and reduce the overall production time. Previous attempts used entropy evaluation metrics which are difficult to translate the terms to the optimising game and a slight misinterpretation often leads to incorrect measurement. This thesis proposes an alternative method which evaluates generated game data with reference result from the testers. The thesis first presents a reliable way to extract information for NPCs classification called Relative Region Feature (RRF). RRF provides an excellent data compression method, a way to effectively classify, and a way to optimise objective-oriented adaptive NPCs. The formalised optimisation is also proved to work on classifying player skill with the reference hall-of-fame scores. The demonstration are done on the on-line competition version of Ms PacMan. The generated games from participating entries provide challenging optimising problems for various evolutionary optimisers. The thesis developed modified version of CMA-ES and PSO to effectively tackle the problems. It also demonstrates the adaptivity of MCTS NPC which uses the evaluation method. This NPC performs reasonably well given adequate resources and no reference NPC is required.