In this study, a closed-loop controller for chest compression which adjusts chest compression depth according to the coronary perfusion pressure (CPP) was proposed. An effective and personalized chest compression method for automatic mechanical compression devices was provided, and the traditional and uniform chest compression standard neglecting individual difference was improved. This study rebuilds Charles F. Babbs human circulation model with CPP simulation module and proposes a closed-loop controller based on a fuzzy control algorithm. The performance of the fuzzy controller was evaluated and compared to that of a traditional PID controller in computer simulation studies. The simulation results demonstrated that the fuzzy closed-loop controller produced shorter regulation time, fewer oscillations and smaller overshoot than those of the traditional PID controller and outperforms the traditional PID controller in CPP regulation and maintenance.