The invention discloses a driving style recognition model training method, device and equipment and a driving style recognition method, device and equipment, and belongs to the technical field of automobiles and the technical field of deep learning. The method comprises the following steps: acquiring sample driving data; the sample driving data comprises a vehicle speed mean value, a longitudinal acceleration standard deviation, a steering wheel angle maximum value, a yaw velocity maximum value and an accelerator pedal position change rate signal; clustering the sample driving data, and determining label data of the sample driving data according to a clustering result; and according to the sample driving data and the label data of the sample driving data, training an initial style recognition model to obtain a driving style recognition model. According to the technical scheme, the accuracy of driving style recognition can be improved.
本发明公开了一种驾驶风格识别模型的训练和驾驶风格识别方法、装置及设备,属于汽车技术领域和深度学习技术领域。该方法包括:获取样本驾驶数据;所述样本驾驶数据包括车速均值、纵向加速度标准差、转向盘角度最大值、横摆角速度最大值和加速踏板位置变化率信号;对所述样本驾驶数据进行聚类,并根据聚类结果确定所述样本驾驶数据的标签数据;根据所述样本驾驶数据和所述样本驾驶数据的标签数据,对初始风格识别模型进行训练,得到驾驶风格识别模型。通过上述技术方案,能够提高驾驶风格识别的准确率。
Driving style recognition model training and driving style recognition method, device and equipment
驾驶风格识别模型的训练和驾驶风格识别方法、装置及设备
2024-01-12
Patent
Electronic Resource
Chinese
Driving style recognition method and device, model training method and device and computer equipment
European Patent Office | 2024
|Privacy preserving driving style recognition
IEEE | 2015
|