混合交通环境下逼近最优换道策略的无人车驾驶模型  

Unmanned Vehicle Model with Approximately Optimal Change Strategy under Mixed Traffic Environment

作　　者：路云龙[1] 李文钰[1] 杨月婷[1] Lu Yunlong;Li Wenyu;Yang Yueting(Mathematics and Statistics School of Beihua University,Jilin 132013,China)

机构地区：[1]北华大学数学与统计学院,吉林吉林132013

出　　处：《北华大学学报（自然科学版）》2018年第5期662-668,共7页Journal of Beihua University（Natural Science）

基　　金：吉林省科学技术发展项目(20180519011JH);吉林省教育科学规划项目(GH170093);吉林省大学生创新创业项目(201811923120);北华大学大学生创新项目(201317122;201317118)

摘　　要：针对无人车在现实交通流中的驾驶行为以及车辆间相互影响机制还不明确的现状,提出逼近最优换道策略的无人车驾驶模型.结合无人车周围的交通环境,引入驾驶行为指标体系,利用层次分析法测算指标权值.基于欧氏距离与灰色关联分析的Topsis(Technique for Order Preference by similarity to an Ideal Solution)法建立驾驶行为决策模型来计算换道最优逼近值,代替随机换道策略驾驶模型中的随机概率值;建立逼近最优换道策略的无人车驾驶模型,利用美国洲际5号公路的实际交通数据对新模型进行数值仿真.结果表明:逼近最优换道策略的无人车驾驶模型明显优于随机换道的无人车驾驶模型,能够改善交通拥堵状况,提高整个道路车辆的行驶速度.Based on the cellular automata,we establish the unmanned vehicle driving model under the optimal approximating lane-changing strategy to analyze the driving behavior of unmanned vehicles in real traffic flow and the interaction mechanism between vehicles. Using the hierarchical analysis method,the TOPSIS（ Technique for Order Preference by similarity to an Ideal Solution） method with the gray relational analysis,a driving behavior decision model is built to calculate the optimal approximating value of land-changing. Using the real traffic data of the American Intercontinental Highway 5 for simulating the new driving model,it is shown that the unmanned vehicle driving model with the optimal lane-changing strategy is superior to the unmanned vehicle model with the random changing strategy. It can improve the traffic jam and improve the speed of vehicle in the whole road.

关 键 词：无人驾驶车 元胞自动机 混合交通流 逼近最优换道策略 

分 类 号：U491[交通运输工程—交通运输规划与管理]