In this paper, visual perception of frontal view in intelligent cars is considered. A Pan-Tilt-Zoom (PTZ) camera is used to track preceding vehicles. The aim of this work is to keep the rear view image of the target vehicle stable in scale and position. An efficient real time tracking algorithm is integrated. It is a generic and robust approach, particularly well suited for the detection of scale changes. The camera rotations and zoom are controlled by visual servoing. The methods presented here were tested on real road sequences within the VELAC demonstration vehicle. Experimental results show the effectiveness of such an approach. The perspectives are in the development of a visual sensor combining a PTZ camera and a standard camera. The standard camera has small focal length and is devoted to an analysis of the whole frontal scene. The PTZ camera gives a local view of this scene to increase sensor range and precision.