论文部分内容阅读
针对驾驶人在换道时若出现决策失误,极易引发交通事故的问题,通过在真实交通环境中进行实车试验,采集车辆运动状态、驾驶人操作行为以及头部运动特性、周围交通环境等数据;通过对换道意图阶段和车道保持阶段数据参数的对比分析,提取能够表征驾驶人换道意图和行为的特征参数;通过建立BP神经网络模型,以不同特征参数作为输入向量对待测样本进行预测,确定最终的输入特征指标,并基于建立的BP网络模型,进行驾驶人换道行为预测。研究结果表明:换道前2s内的预测准确率为94.4%,灵敏度为93.33%,能够准确预测出93.33%的换道行为;该模型能够有效预测驾驶人的换道行为,且准确率高、时序性强。
Aiming at the driver’s mistake in decision-making when changing lane, it can easily lead to the traffic accident. Through real vehicle test in real traffic environment, the vehicle movement status, driver’s operation behavior, head movement characteristics, surrounding traffic environment, etc. are collected Through the comparison and analysis of the data parameters in lane changing phase and lane keeping phase, the characteristic parameters which can characterize driver lane changing intention and behavior are extracted. By establishing BP neural network model, different characteristic parameters are taken as input vector Forecast, determine the final input characteristic index, and based on the established BP network model, make driver lane changing behavior prediction. The results show that the accuracy of prediction is 94.4% and the sensitivity is 93.33% within 2 seconds before lane change, and 93.33% lane change behavior can be accurately predicted. This model can effectively predict driver lane changing performance with high accuracy, Timing is strong.