The problem of finding a finite planar emulator of a non-projective graph, or proving that it does not exist, is without any doubt extremely difficult. So far, there is no known universal method that would guarantee obtaining a result (not even by brute force). Student's task is to explore the option of using evolutionary computations, i.e. genetic algorithms and genetic programming, for constructing finite planar emulators of graphs. Part of the problem is to propose a fitness measure of planarity that would be suitable for use within such techniques.