基于隐Markov的机动车驾驶人状态预测

Prediction of Automation Driving State Based on Hidden Markov Model

以后车车速、前后车速差和车间距作为观察变量输入,驾驶人的驾驶状态作为隐含变量输出,利用隐马尔科夫模型提出了一种机动车驾驶人状态预测方法。首先筛选预测所需的观察状态序列,接着利用隐Markov求解预测时刻所有选中的观察状态序列出现的概率以及各观察状态序列和指定驾驶状态(隐含变量)同时出现的概率,最后利用条件概率将上述两者转化为驾驶人状态概率。为检验方法的预警性能,除考核预报的正确性外,定义了"预报度"衡量驾驶人不良状态概率为P时提前预警的时间。仿真结果表明,预测的驾驶状态变化趋势与PERCLOS监测结果一致,且能实现提前预报,P值越小,预报度越大。

Based on Hidden Markov Model, a new prediction method on driving state is advanced. In which, the velocity of following car, the velocity difference and distance headway is input as observation variables, the driving state is output as hidden variable. First the set of observation state is classified, then the probability of observation states need is calculated by forward algorithm, and the probability of observation states and driving state appeared together is calculated, at last, the prediction value of driving state could be got by conditional probability. The warning character of the prediction method could be evaluated not only by accuracy but also by a new index ＂predictability advanced＂, which could show the degree of warning time at p probability. The results of simulation show that the method is right and could predict the driving state advanced, in which the smaller the value P is, the more the degree of predictability is.