计算机与数字工程2019，Vol.47Issue(8)：1930-1934,5.DOI:10.3969/j.issn.1672-9722.2019.08.018
降低车道偏离预警系统误报率方法研究
Research on the Method of Reducing the False Alarm Rate of LDWS
摘要
Abstract
For lane departure warning system(LDWS)cannot effectively distinguish between lane and unconscious lane de?parture problem,the data of vehicle lane changing and lane changing are obtained by real vehicle experiment,the steering wheel an?gle,the lateral velocity and the transverse distance are extracted as the identification parameters of the model,and the extended Kal?man filter,normalized and K mean clustering method are used to process the data. Support vector machine(SVM)is used to identi?fy lane departure behavior. In order to improve the recognition rate of SVM,particle swarm optimization(PSO)algorithm is used to optimize the SVM parameters. The results show that the recognition rate of the optimized model is more than 88% when the time win?dow is 3.5s,which can meet the application requirements of LDWS system.关键词
车道偏离/卡尔曼滤波/支持向量机/粒子群算法Key words
lane departure/Kalman filter/SVM/PSO分类
信息技术与安全科学引用本文复制引用
孟妮,山岩..降低车道偏离预警系统误报率方法研究[J].计算机与数字工程,2019,47(8):1930-1934,5.基金项目
国家自然科学基金项目(编号:61374196) （编号:61374196）
陕西省自然科学基金项目(编号:2016JQ5096)资助. （编号:2016JQ5096）
