论文部分内容阅读
为准确辨识驾驶员酒驾行为以及酒驾状态水平,提高酒驾治理效率,通过人因工程试验和驾驶模拟试验,采集并预处理驾驶员在正常、饮酒、醉酒3种驾驶状态下的驾驶行为数据(包括驾驶员的人、车、环境数据);对原始参数进行因子分析,提取特征参数并将其作为多层神经网络的输入向量,训练多层神经网络,建立基于因子分析和多层神经网络的酒驾行为辨识模型;选取75组测试样本数据输入模型,将模型的输出结果与实际情况比较,验证模型的有效性。研究表明:该模型的训练时间为0.905 s,最优验证均方误差(MSE)为0.034,识别准确率达92.41%,用该模型能较为快速、准确地识别酒后驾驶行为。
In order to accurately identify the driver’s drunk driving behavior and the drunk driving status, improve the efficiency of drunk driving management, and collect and preprocess the driver’s driving behavior data under normal driving, drinking and drunk driving conditions through human engineering test and driving simulation test Driver’s person, vehicle, environment data); factor analysis of the original parameters, extraction of the characteristic parameters and as a multi-layer neural network input vector, training multi-layer neural network, the establishment of based on factor analysis and multilayer neural network drunk driving Behavioral identification model; select 75 sets of test sample data input model, the output of the model compared with the actual situation to verify the effectiveness of the model. The research shows that the training time of this model is 0.905s, the MSE is 0.034, and the recognition accuracy is 92.41%. Using this model, the drunk driving behavior can be identified more quickly and accurately.