Issues associated with the use of a visual sensor in the feedback loop are addressed. In particular, algorithms are proposed for the solution of the robotic (hand-eye configuration) visual tracking and servoeing problem. The problem of robotic visual tracking is stated as a problem of combining control with computer vision. The use of sum-of-squared differences (SSD) optical flow for the computation of the vector of discrete displacements is proposed. These displacements are fed to an adaptive controller (self-tuning regulator) that drives the robot in conjunction with a Cartesian robotic controller. Three different adaptive control schemes have been implemented, and the results are presented.