This paper presents a procedure for the design of a fuzzy logic controller for the garaging of a wheeled mobile robot, with non-holonomic constraints and discrete control signal levels. The design procedure is based on the virtual fuzzy magnet principle, which implies the definition of fictive points in the garage surroundings, and which, in an appropriate manner, by robot attraction, provides an efficient garaging process. The proposed fuzzy logic controller has four input variables: two represent the robot's distancing from fictitious fuzzy magnets and two are angles that define the orientation of the vehicle. The output variables are related to voltages sent to motors in charge of propelling the left and right wheel of the robot. The efficiency and shortfalls of the proposed algorithm are analyzed by means of both detailed simulations and multiple re-runs of a real experiment. Special attention is devoted to the analysis of different initial robot configurations and the effect of an error in the estimation of the current position of the robot on garaging efficiency.