基于DBN的特种车辆前向防撞推理模型  

Forward Collision Avoidance Reasoning Model of Special Vehicles Based on DBN

作　　者：仝兆景[1] 张艳杰[1] 赵运星[1] 芦彤 TONG Zhao-jing;ZHANG Yan-jie;ZHAO Yun-xing;LU Tong(School of Electrical Engineering and Automation,Henan Polytechnic University,Jiaozuo 454000,China)

机构地区：[1]河南理工大学电气工程与自动化学院

出　　处：《测控技术》2019年第10期56-60,共5页Measurement & Control Technology

基　　金：国家自然科学基金项目资助(U1504623);河南省高校基本科研业务费专项资金资助(NSFRF1615);国家安全监管总局2017安全生产重特大事故防治关键技术科技项目(Henan-0008-2017AQ);河南省高等学校矿山信息化重点学科开放实验室开放基金项目(KY2015-07)

摘　　要：针对特种车辆在动态环境中的前向碰撞风险评估问题,对特种车辆前向碰撞风险的自然因素、驾驶员行为特征等进行研究,并对固定的车辆安全防撞距离阈值进行改进,提出了一种基于动态贝叶斯网络的前向防撞推理模型。该模型将自车与周围环境的位置关系、环境关系、驾驶员行为等因素进行融合,一旦周围环境发生变化,该模型可以及时评估前向风险,并与静态贝叶斯网络的前向推理模型进行对比分析。仿真实验验证了该前向防撞推理模型的可行性和有效性。In order to solve the problem of the forward collision risk assessment of special vehicles in dynamic environment,the natural factors of forward collision risk of special vehicles and the driver’s behavior characteristics were studied,and the fixed threshold of vehicle safety collision distance was improved,a forward collision avoidance reasoning model based on dynamic Bayesian network was proposed.The model integrates the relationships between the vehicle and the location of the surrounding environment,the environment,driver behavior and other factors.Once the surrounding environment changes,the model can evaluate the forward risk in time,and compare it with the forward reasoning model of static Bayesian network.The simulation experiments verify the feasibility and effectiveness of the forward collision avoidance reasoning model for special vehicles with dynamic Bayesian networks.

关 键 词：特种车辆 动态贝叶斯网络 静态贝叶斯网络 前向防撞 

分 类 号：TP277[自动化与计算机技术—检测技术与自动化装置]