基于混合模糊多人多目标非合作博弈的车道选择模型

发布时间：2018-02-14 00:49

本文关键词： 智能交通系统驾驶倾向性车辆集群态势多人多目标对策混合模糊对策　出处：《自动化学报》2017年11期 　论文类型：期刊论文

【摘要】：建立汽车安全驾驶辅助系统(包括安全驾驶预警系统)是保证交通安全的有效手段.准确预测车辆集群态势是汽车安全辅助驾驶的前提,车道选择是车辆集群态势发生转移最为根本的原因,也是交通流理论研究的基本内容.以往研究没有综合考虑车辆集群复杂态势下各运动实体特征及其操控者类型,以及多个车道间车辆的冲突对车道选择的影响.为此,本文综合考虑各运动实体特征及其操控者类型,基于混合模糊多人多目标非合作博弈方法,建立城市快速路基本路段上的驾驶员车道选择模型.通过分析各方驾驶员在不同车道选择策略下的收益,确定换道博弈的Nash均衡,得到驾驶员最优车道选择策略.研究结果表明:基于混合模糊多人多目标非合作博弈方法建构的驾驶员车道选择模型,其预测准确率可达到85.2%.
[Abstract]:The establishment of vehicle safety driving assistance system (including safe driving warning system) is an effective means to ensure traffic safety. Accurate prediction of vehicle cluster situation is the premise of vehicle safety auxiliary driving. Lane selection is the most fundamental reason for the movement of vehicle cluster situation, and it is also the basic content of the research on traffic flow theory. Previous studies have not considered the characteristics of each movement entity and the type of operator in the complex situation of vehicle cluster. And the impact of vehicle conflict on lane selection. Therefore, considering the characteristics of each movement entity and the type of operator, this paper based on the hybrid fuzzy multi-objective non-cooperative game method. This paper establishes the driver lane selection model on the basic section of urban expressway. By analyzing the income of drivers under different lane selection strategies, this paper determines the Nash equilibrium of the switch game. The optimal driveway selection strategy is obtained. The results show that the prediction accuracy of the model based on mixed fuzzy multi-objective non-cooperative game method can reach 85.2.
【作者单位】：山东理工大学交通与车辆工程学院;清华大学汽车安全与节能国家重点实验室;
【基金】：汽车安全与节能国家重点实验室开放基金(KF16232) 山东省自然科学基金(ZR2014FM027,ZR2017LF015) 山东省社会科学规划研究项目(14CGLJ27) 国家自然科学基金(61074140,61573009,51508315,51608313) 山东省高等学校科技计划(J15LB07)资助~~
【分类号】：O225;U463.6;U495

【相似文献】