论文部分内容阅读
快速城镇化造成的交通拥堵、交通安全、交通环境等问题日益突出,从交通参与者个体的特性出发,分析驾驶风格多样性,从而建立“以人为本”的智能交通系统,是充分开发利用现有道路交通资源的有效途径。然而,驾驶风格难以检测和量化,这导致个体特征与系统性的计算难以进一步融合。该文通过实验车采集了16位驾驶员在实际道路上的驾驶行为数据,利用主题模型挖掘驾驶行为中的隐含主题,将数据结构由“驾驶风格—驾驶行为数据”转化为“驾驶风格—驾驶状态—驾驶行为数据”结构,发现了驾驶风格结构化信息,能够为建立更为有效的智能交通系统提供科学依据与理论支持。通过分析相关性,证实了模型重构数据与原数据有较好的一致性,验证了模型进行驾驶风格多样性分析的可行性。
The problems of traffic jam, traffic safety and traffic environment caused by rapid urbanization have become increasingly prominent. Based on the characteristics of individual traffic participants, the analysis of driving style diversity and the establishment of a “people-oriented” intelligent transportation system are essential measures for developing and utilizing the existing There is an effective way of road traffic resources. However, driving style is difficult to detect and quantify, which makes it difficult to further integrate individual characteristics with systematic calculations. This article collected the driving behavior data of 16 drivers on the actual road through the experimental vehicle, and used the theme model to excavate the hidden topics in the driving behavior, and converted the data structure from “driving style - driving behavior data ” into Driving style - driving behavior - driving behavior data "structure, and found that driving style structured information can provide scientific basis and theoretical support for establishing a more effective intelligent transportation system. By analyzing the correlation, it is confirmed that the reconstructed data of the model is in good agreement with the original data, which verifies the feasibility of the model for driving style diversity analysis.