In this thesis, the target tracking system for an intelligent wheelchair using infrared range-finder and CCD camera is discussed. The tracking system is designed such that the wheelchair can follow a path of human once in a while. The problem of human tracking requires a recognition of feature as well as a tracking of human position. Therefore, infrared range-finder and CCD camera create target candidate through each target recognition algorithm. But, target tracking that uses single vision sensor has low accuracy due to uncertainties of target decision and position error. In this thesis, therefore, a data fusion algorithm which reduces the uncertainties of target decision and corrects the position error of human is proposed. The effectiveness of the proposed method is verified through experiments.