Learning of robot movement

U ovom radu opisan je princip rada ojačanog učenja, odnosno Q učenja te njegova implementacija na agenta. Promatrano je više različitih algoritama učenja agenta te su istraženi njihovi utjecaji na promjenu gibanja agenta.In this thesis, the principle of machine learning (with focus on Q-learning) is described. Multiple algorithms that make an agent learn in different ways have been observed and their influences on the agent's motion have been recorded