高级车辆事故自动呼救（Advanced Automatic Crash Notification，AACN）系统能在车辆发生碰撞事故时及时根据驾驶员伤情预测算法预测车内驾驶员伤情，有助于救援中心做出早期判断并制定积极有效的救援方案，从而提高救援效率，挽救更多重伤驾驶员的生命。选取速度变化量、事故碰撞方向、驾驶员年龄、性别、是否佩戴安全带以及驾驶员侧安全气囊是否打开作为引起驾驶员伤情的影响因素；利用道路事故数据分析并构建Logistic回归模型，使用Hosmer-Lemeshow测试表验证了模型的有效性，并通过敏感性分析获得了最佳触发阈值。然后，提出了一种驾驶员伤害预测算法，并基于该算法实现了AACN系统终端的总体设计。最后，通过一个实际案例来检验伤情预测算法的准确性。结果表明，提出的驾驶员伤情预测算法准确率较高，能够有效预测驾驶员伤情，也提高了AACN系统的准确性。
When a vehicle collision occurs, the Advanced Automatic Crash Notification (AACN) system can predict the driver’s injury in the vehicle based on the driver’s injury prediction algorithm. It helps the rescue center make early judgments as well as positive and effective decisions, which improves rescue efficiency and saves more lives of seriously injured drivers. First, by selecting the amount of speed change, the direction of the accident, the driver’s age, gender, whether to wear the seat belt, and whether the driver’s side airbag inflated as the influencing factors of the driver’s injury, a Logistic regression model was analyzed and built based on road accident data. Next, the effectiveness of the model was verified by using the Hosmer-Lemeshow test table, and the best trigger threshold was obtained through sensitivity analysis. Then, a driver injury prediction algorithm was proposed, and the overall design of the AACN system terminal was realized based on this algorithm. Finally, an actual case was used to test the accuracy of the injury prediction algorithm. The results of the case study show that the proposed driver injury prediction algorithm has a high accuracy rate, can effectively predict the driver’s injury, and improve the accuracy of the AACN system.