The invention discloses a method for detecting distraction of a driver based on a model LSTM-NN (Long Short-Term Memory Neural Network). The method comprises the following steps: 1, acquiring longitudinal and transverse parameters of a vehicle through a signal acquisition module, wherein the longitudinal acquisition parameters comprise a longitudinal speed, a longitudinal acceleration, a vehicle following distance and a speed relative to a front vehicle, and the transverse acquisition parameters are transverse acceleration and a steering wheel angle; 2, preprocessing various index data acquired by a signal acquisition unit through a driving distraction signal detection unit, analyzing and judging the transmitted multiple indexes by a driving distraction behavior judgment unit to obtain the attention state of a driver, and transmitting a result to a distraction early warning module; and 3, enabling a distraction early warning module to determine whether to connect the branch through a triode according to the behavior characteristics of driving distraction so as to prompt the attention state of the driver. The model can learn time series information before and after learning and reliably detect distraction behaviors.
本发明公开了一种基于模型长短时记忆神经网络LSTM‑NN检测驾驶人驾驶分心的方法,包括以下步骤:步骤1,通过信号采集模块采集车辆的纵向与横向参数;其中,纵向采集参数为:纵向速度,纵向加速度,跟车距离,与前车相对速度;横向采集参数为:横向加速度,方向盘转角;步骤2,通过驾驶分心信号检测单元对于信号采集单元采集到的各类指标数据进行预处理,由驾驶分心行为判别单元对传输过来的多类指标进行分析判断,得出驾驶员注意力状态并将结果传送到分心预警模块;步骤3,分心预警模块根据驾驶分心的行为特征,通过三极管确定是否接通该条支路,进行提示驾驶人注意力状态。本发明模型能够学习前后时间序列信息并可靠地检测分心行为。
Method for detecting distraction of driver based on model LSTM-NN (Long Short-Term Memory Neural Network)
一种基于模型长短时记忆神经网络LSTM-NN检测驾驶人驾驶分心的方法
2021-04-13
Patent
Elektronische Ressource
Chinesisch
IPC: | B60W CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION , Gemeinsame Steuerung oder Regelung von Fahrzeug-Unteraggregaten verschiedenen Typs oder verschiedener Funktion |
Online Driver Distraction Detection Using Long Short-Term Memory
Online Contents | 2011
|