A novel saturated visual control strategy for nonholonomic mobile robots is presented. The principal focus is to deal with the stabilization problem. The saturation bounds for the control actions avoid exceeding the physical restrictions of the actuators. This feature is the main difference concerning related works. The closed-loop system is analyzed using the theory of non-autonomous cascade systems, and global uniform asymptotic stability is guaranteed. Experimental validation is carried out by using a unicycle-type mobile robot with an onboard camera. The experiment consists of stabilizing the robot in a preset pose using visual information. The experimental results are compared with a literature unsaturated visual servoing strategy. The results confirm the effectiveness of the proposed approach, completing the assigned task and bounding the control inputs.