The aim of this coursework was to develop an evolutionary algorithm that would optimise the weights of a multi-neuron percepton for employment in the automated moon landing problem. The included report documents the implementation and parameter tuning of two solutions: a steady state genetic algorithm and an application of the simulated annealing technique.