3^rd and 4^th Year Projects

· Artificial Life

You can model some simple single-celled creatures wich can move around in a virtual world. You can model the internal structure of the creatures i.e. how they convert food into energy and repair damage to themselves. The creatures can get together and cooperate with each other or they can compete against each other, even eating one another. They can evovle by Darwinian evolution. Do they become more complex? Do multi-cellular creatures emerge?

· Virtual Ant Colony

Ants in a colony carry out various activities: collecting food, removing rubbish, defending the colony, looking after the young etc. Each activity is carried out at a different location. The ants leave pheromone trails behind them indicating which activity they are involved in. The ants detect the trails left by other ants and decide to switch activities if they detect too many trails from other ants involved in the same activity. The whole system is spontaneously self-organising and complex structures can emerge. Your task is to model such a system.

· Modelling simple societies - game theory

People in groups behave in different ways, some are cooperative and trusting others are aggressive and deceptive. Your task is to try to create a virtual society in which individuals behave in different ways. Each has a different strategy, which may evolve as time goes on. They individuals may learn and improve their behaviour to obtain greater rewards. What strategy works out best in the long-run?

· Digital Signal Processing applied to de-blurring images

Images can be blurred due to the camera being out of focus or due to motion. Digital signal processing techniques can be applied to de-blur the images. The same techniques can be used to improve the resolution of microscope images. Your task is to acquire some images and see if you can improve them.

· Mobile Robot with Camera

You'll fit a Rug Warrior mobile robot with a small camera and get it to navigate around a real environment. You can insert objects into this environment such as model trees, people, vehicles, traffic lights etc - perhaps even other robots. The robot will have to recognise these other objects and take appropriate action in each case.