Abstract

This paper introduces another approach to pre-crash velocity determination by the means of genetic algorithm model adjustment (GAMA). This innovative method is based on the standard genetic algorithm approach applied to the problem of determining the dependency of predicted variable on given parameters. The experiment is performed on a database obtained from NHTSA (National Highway Traffic Safety Administration), which contains potent number of data samples regarding frontal vehicle crash tests: i.e. vehicle mass, deformation zone and deformation coefficients C1-C6. Various classes of possible functions are considered and the model for determining the pre-crash speed based on the given parameters is obtained.

摘要

本文介绍了另一种通过遗传算法模型调整（GAMA）确定碰撞前速度的方法。这种创新方法基于应用于确定预测变量对给定参数的依赖性问题的标准遗传算法方法。该实验是在从 NHTSA（国家公路交通安全管理局）获得的数据库上进行的，该数据库包含大量关于正面车辆碰撞测试的数据样本：即车辆质量、变形区和变形系数 C1-C6。考虑了各类可能的函数，并获得了基于给定参数确定预碰撞速度的模型。