This week begun by putting all the research I had made the previous week to use. I drafted some simple dissertation ideas and emailed relevant lecturers inquiring about their validity.
I first contacted Dr. Ruth Falconer. She is a Senior Lecturer in Abertay's division of computing and mathematics specializing in visualizing data and procedural content generation. With that in mind, I wrote to Dr. Falconer with my draft dissertation idea.
A topic that had caught my attention was AI steering behaviours (some can be seen here). My fascination was with the seemingly complex behaviours you could achieve using an extremely simple framework per object and my dissertation aim would be to answer, "How can we use this to easily create complex gameplay?"
Dr. Falconer responded promptly with encouragement and subsequent links with more information on the subject such as the red3d article on boids. (this can be found at https://www.red3d.com/cwr/boids/) and relevant research papers that involved these autonomous agents.
I then contacted Dr. David King a Lecturer also part of Abertay's division of computing and mathematics who specializes in the practical application of academic AI techniques such as neural networks and self-organizing maps.
While I approached Dr. King with roughly the same ideas that I presented to Dr. Falconer after scheduling and attending a face to face meeting with Dr. King I formulated a new dissertation approach on how would incorporating more formal AI techniques in those aforementioned steering behaviours affect their performance and results.
Dr. King provided me with numerous avenues I could explore and even shared a previous honours dissertation relating to mine titled "An Evaluation of Fast Multi-Layer Perceptron Training Techniques for Games" which posits for ways to make usage of Neural networks in games.
Having taken both Lecturer's comments and feedback I've set to come up with a clearer dissertation topic and further researching the topic of AI in games.