A comparison between Position Based and Image Based Visual Servoing on a 3 DOFs translating robot