With the introduction to proactive HMI, the vehicle can predict users’ intentions and start functioning, thereby reducing distractions, increasing spirituality, and improving driving safety and user experience. With proactive HMI, the accuracy of the use intention prediction model at the mechanism levels to become the key to affecting the quality of the proactive HMI experience.However, there is lack of effective means for improving the accuracy of the user prediction model,and there is not sufficient relevant research. Intelligent interaction is an effective method to improve the performance of robots. By introducing this technology into the design of active responsive interaction, users’ intentions can be obtained to help to break for the bottlenecks of current algorithms.This paper proposes a framework of automobile active responsive interaction design based on smart interaction and using intelligent interaction to improve prediction accuracy, the key points that deserve attention are also listed, and the specific design case is illustrate.