基于模糊支持向量机的旁车道车辆并线意图识别方法  

Identification of Cut-in Maneuver of Side Lane Vehicles Based on Fuzzy Support Vector Machines

作　　者：马国成[1] 刘昭度[1] 裴晓飞[1] 王宝锋[1] 齐志权[1] 

机构地区：[1]北京理工大学机械与车辆学院,北京100081

出　　处：《汽车工程》2014年第3期316-320,共5页Automotive Engineering

基　　金：国家自然科学基金(51005019)资助

摘　　要：为检测旁车道车辆驾驶员的并线意图,利用机器学习技术基于模糊支持向量机建立了并线意图识别器。识别器的训练样本由实际交通环境中的车辆并线数据获得,包括主车道与旁车道车辆的7个运动属性,其中对不能直接利用传感器信息获取的属性由Kalman滤波器预估得到。由于在并线初始时刻的并线样本不能有效区别于非并线样本,所以在支持向量机的求解中引入样本模糊隶属度系数以提高并线意图识别器训练的准确性,同时对支持向量机中的参数基于交互检验正确率进行网格优化。在实际交通环境中对并线意图识别器进行了试验,结果表明,识别器工作有效,经过简单处理后的识别结果可有效反映驾驶员的并线意图。For detecting the driver＇ s cut-in intention of a side lane vehicle,a cut-in intention identifier is built using machine learning technique based on fuzzy support vector machines （FSVM）.The training samples of identifierobtained from the cut-in data of vehicles in real traffic environment possess 7 motion attributes of vehicles in both main lane and side lane,in which the attributes not able to get directly from sensors are pre-estimated by Kalman filter.Due to the cut-in samples cannot be effectively distinguished from non cut-in samples at the initial moment of cut-in,a fuzzy membership coefficient is introduced for each sample in solving FSVM to improve the training accuracy of cut-in identifier,and a grid optimization is conducted on the parameters of FSVM aiming at the highest correctness rate of cross validation.The results of the test on cut-in intention identifier in real traffic environment show that the identifier is effective in work with identification results effectively reflect the cut-in intention of driver after simple processing.

关 键 词：驾驶员辅助系统 并线意图识别 模糊支持向量机 

分 类 号：U463.6[机械工程—车辆工程]