Car-following models, which describe the interaction between successive vehicles in the same lane, have been studied for decades. A group of models are derived from the stimulus-response pattern concentrating on the effect of diverse stimulus type. This study presents a potential-based car-following model using the concept of the artificial potential field, which aims for the precise and fast interactive operations in an evolving environment. Spacing headway is divided into two parts according to the potential influence region. The variation rate of the spacing headway generates the control force within the potential influence region, while the difference of desired and current velocity takes control out of the influence range. Calibration and validation of the simplified model are conducted using NGSIM data. Statistical tests show that the proposed model can reproduce the car-following process very well.