In our project on the autonomous guidance of Micro-Air Vehicles (MAVs) in confined indoor and outdoor environments, we have developed a vision based autopilot, with which a miniature hovercraft travels along a corridor by automatically controlling both its speed and its clearance from the walls. A hovercraft is an air vehicle endowed with natural roll and pitch stabilization characteristics, in which planar flight control systems can be developed conveniently. Our hovercraft is fully actuated by two rear and two lateral thrusters. It travels at a constant altitude (∼2 mm) and senses the environment by means of two lateral eyes that measure the right and left optic flows (OFs). The visuo-motor control system, which is called LORA III (Lateral Optic flow Regulation Autopilot, Mark III), is a dual OF regulator consisting of two intertwined feedback loops, each of which has its own OF set-point and controls the vehicle's translation in one degree of freedom (surge or sway). Our computer-simulated experiments show that the hovercraft can navigate along a straight or tapered corridor at a relatively high speed (up to 1 m/s). It also reacts to any major step perturbations in the lateral OF (provided by a moving wall) and to any disturbances caused by a tapered corridor. The minimalistic visual system (comprised of only 4 pixels) suffices for the hovercraft to be able to control both its clearance from the walls and its forward speed jointly, without ever measuring speed and distance. The non-emissive visual sensors and the simple control system developed here are suitable for use on MAVs with a permissible avionic payload of only a few grams. This study also accounts quantitatively for previous ethological findings on honeybees flying freely in a straight or tapered corridor.