ISSN 1674-8484 汽车安全与节能学报, 第 13 卷 第 1 期 , 2022 年 10/21 CN 11-5904/U J Automotive ns of following car dangerous driving or front car emergency braking, through a driving simulation platform with a simulation software of Virtual Test Driving (VTD). The relative distances, the relative speeds, the decelerations of the preceding car, the collision times, and the lateral distances were taken as the input parameters. Model mobility was verified by an untrained sample data, and compared with a traditional Back Propagation (BP) 收稿日期 / Received ： 2021-05-30。 修回日期 / Revised ： 2021-10-20。 基金项目 / Supported by ： 国家重点研发计划项目 (2017YFB0102500) ；天津市人工智能科技重大专项 (17ZXRGGX00070) ；天津 市科技计划项目 (21YDTPJC00350) ；天津职业大学科学研究基金项目 (20181109)。 第一作者 / First author ： 房亮 (1988 － )，男 ( 汉族 )，河北，讲师。 E-mail: ******@。 * 通讯作者 / Corresponding author ： 关志伟，教授。 E-mail: ******@。房亮，等：基于深度学习 LSTM 的智能车辆避撞模型及验证 105 neural network. The results show that the R2 values of the model are higher than that of the traditional BP model by and , respectively for the predicational decelerations and the predicational steering wheel angles. Therefore, this model has a fitting goodness for the big data s
