计算机与数字工程2017，Vol.45Issue(12)：2336-2340,5.DOI:10.3969/j.issn.1672-9722.2017.12.003
基于神经网络算法的车辆行驶识别研究
Research on Vehicle Identification Based on Neural Network Algorithm
摘要
Abstract
Aiming at the identification problem of vehicle driving,by selecting 6 typical road conditions as the initial condi?tions for each sample identification,a typical road conditions are divided into blocks to identify the sample expansion,and selected 10 to fully characterize the working characteristic and calculating parameters,characteristic parameters of every kind of typical the condition of value standard parameter vector is normalized after the formation of the corresponding neural network mode recognition model,constructing learning vector quantization of the initial condition of recognition model for effective training in order to improve the precision of the models.The neural network algorithm is completed after the training of the model,and the work condition identi?fication and simulation test are carried out under the condition of comprehensive test.The experimental results show that the trained neural network algorithm can effectively identify the actual working conditions.关键词
神经网络/学习向量量化/车辆行驶/工况识别Key words
neural network/learning vector quantization/vehicle running/condition identification分类
信息技术与安全科学引用本文复制引用
史骏..基于神经网络算法的车辆行驶识别研究[J].计算机与数字工程,2017,45(12):2336-2340,5.基金项目
国家自然科学基金- 青年项目“基于驾驶意图的电动汽车电液复合制动协调控制系统研究“(编号:51507013)资助. （编号:51507013）
