基于改进灰靶理论的驾驶模拟器人-车-路-环系统评价  

Evaluation of Human-Vehicle-Road-Environment System in Driving Simulator Based on Improved Grey Target Theory

作　　者：陈亮[1] 李明丽 张森 刘拥华[1] 方杰 刘恒辰 CHEN Liang;LI Mingli;ZHANG Sen;LIU Yonghua;FANG Jie;LIU Hengchen(Faculty of Transportation Engineering,Kunming University of Science and Technology,Kunming 650500,China;Yunnan Research Institute of Highway Science and Technology,Kunming 650000,China)

机构地区：[1]昆明理工大学交通工程学院,云南昆明650500 [2]云南省公路科学技术研究院,云南昆明650000

出　　处：《昆明理工大学学报(自然科学版)》2025年第2期160-172,共13页Journal of Kunming University of Science and Technology(Natural Science)

基　　金：国家自然科学基金项目(72261021).

摘　　要：汽车驾驶模拟系统已广泛应用于交通领域的科学研究,模拟系统有效性高低是其应用发展的重要前提.驾驶模拟系统整体有效性研究是反映其应用可靠性状态的必要手段,为综合考虑道路交通人-车-路-环系统对于驾驶模拟系统的有效性,本研究以模拟系统为中心,分析“车”“人-车”“车-路”“人-车-路-环”交互影响有效性典型要素,基于人机工效、车辆动力学、驾驶感知等理论构建评价指标体系;鉴于评价指标体系繁多复杂的特点,本研究采用灰靶理论构建评价模型,并通过改进粒子群算法提高传统灰靶模型的分辨能力与评价结果精确性;选取国内某驾驶模拟系统实例,搭建真实和虚拟实验场景,获取评价参数数据,以理论值或实车测量值为校标,模拟系统参数值为模型输入,得到该模拟系统综合有效性为中等,其中人机工效有效性良好,图文视认有效性较差,车辆动力学、道路感知、风险感知有效性为中等;结合各评价主观分析表明,评价结果基本一致.本研究为模拟系统综合有效性等方面研究及优化提供思路与方法,助力驾驶模拟系统在智能交通等领域的应用.Automotive driving simulation systems have been widely used in scientific research in the transportation field,and the effectiveness of these systems is a critical prerequisite for their application and development.Research on the overall effectiveness of driving simulation systems is essential to reflect their reliability in practical applications.To comprehensively consider the impact of the human-vehicle-road-environment system on the effectiveness of driving simulation systems,this study focuses on the simulation system and analyzes typical factors affecting the effectiveness of interactions in“vehicle”,“human-vehicle”,“vehicle-road”,and“human-vehicle-road-environment”systems.An evaluation index system is constructed based on theories of human-machine ergonomics,vehicle dynamics,and driving perception.Given the complexity and multiplicity of the evaluation index system,the grey target theory is adopted to build an evaluation model,and the traditional grey target model is improved using a particle swarm optimization algorithm to enhance its discriminative ability and evaluation accuracy.A domestic driving simulation system is selected as a case study,and real and virtual experimental scenarios are established to collect evaluation parameter data.The simulation system parameters are compared with theoretical values or actual vehicle measurements.The results show that the overall effectiveness of the simulation system is moderate,with good effectiveness in human-machine ergonomics,poor effectiveness in graphic recognition,and moderate effectiveness in vehicle dynamics,road perception,and risk perception.Subjective analysis of each evaluation aspect indicates that the evaluation results are consistent.This study provides insights and methods for research and optimization of the comprehensive effectiveness of simulation systems,supporting the application of driving simulation systems in fields such as intelligent transportation.

关 键 词：交通工程 驾驶模拟系统 有效性 灰靶理论 粒子群算法 

分 类 号：U491.25[交通运输工程—交通运输规划与管理]