摘 要：驾驶风格作为影响汽车性能表现的关键因素之一，不仅直接关系到车辆的能耗水平、续航里程，还深刻影响道路安全及交通流畅度。本文提出了一种基于改进K均值聚类的驾驶风格自动识别算法。在利用因子分析对实车上传的周期数据降维后，使用该算法能够有效识别驾驶风格，并对结果进行有效解读与分析。结果表明，驾驶风格可分为平静型、保守型与激进型三类。
关键词：改进K均值聚类；因子分析；驾驶风格识别
中图分类号：U461.91；TP301.6 文献标志码：A DOI：10.15917/j.cnki.1006-3331.2025.04.001
Research on Automatic Recognition Algorithm of Automobile Driving Style Based on Improved K-Means Clustering Analysis
CHEN Zhen,LIANG Fengshou,LU Gaolin,ZHANG Luchan,CHEN Peng
Abstract: As a key factor affecting commercial vehicle performance,driving style significantly impacts not only vehicle energy consumption and range but also road safety and traffic fluency.This paper proposes a driving style recognition algorithm based on an improved K-means clustering approach.After reducing the dimension of periodic real vehicle operational data using factor analysis,the algorithm effectively identifies driving styles and interprets and analyzes the results.The results show that driving styles can divided into three distinct types:calm,conservative,and aggressive.
Key words: improved K-means clustering; factor analysis; driving style recognition
