基于脑电疲劳监测的车辆避撞预测模型研究

Research on vehicle collision avoidance prediction model based on EEG fatigue monitoring

为了减少交通事故的发生,解决传统固定阈值模型带来的预警过早或过晚问题,本文综合考虑了人在驾驶中容易出现因脑疲劳而注意力不集中的状况,提出了新的车辆避撞预测模型。研究中引入大脑疲劳影响因素,实验分析了人脑疲劳的脑电信号,引入节律能量值,建立了脑电疲劳监测的车辆碰撞预测模型,修正了传统的安全碰撞时间。实时获取车辆的运动状态和目的位置坐标,对比实时获取车辆的安全避撞时间和实际碰撞时间,在出现危险时,提醒驾驶员及时采取制动措施。最后,在不同车速情况下,计算最佳安全避撞时间,验证了所提脑电疲劳预测模型的有效性。

With the frequent occurrence of traffic accidents,a new vehicle collision avoidance prediction model is proposed to solve the problem of early or late warning of the traditional fixed threshold model.This work comprehensively considers that people tend to be distracted due to brain fatigue in driving.The influence factors of brain fatigue is introduced,the EEG(electroencephalogram)signal of brain fatigue is analyzed,the rhythm energy value is introduced,the vehicle collision prediction model of EEG fatigue monitoring is established,and the traditional safe collision time is revised.Then,the vehicle’s motion state and target position coordinates are acquired in real time.The safety collision avoidance time and actual collision time of the vehicle are compared in real time.In case of danger,the driver is reminded to take braking measures in time.Finally,the optimal safe collision avoidance time is calculated under different vehicle speeds to verify the proposed computational EEG fatigue prediction model.