摘要
针对当前'请求-响应'被动式行车服务存在的个性化、人性化、实时性等不足,提出了基于多智能体(multi-agent system,MAS)分层集中控制的'自动识别-主动推送'形式的行车主动服务系统(driving active service system,DASS).DASS将行车环境实体与服务作为建模对象,提出基于规划图的行车服务需求模型及用户需求与服务能力匹配算法.以交通安全预警和实时动态路径规划为例,通过自主搭建的驾驶员在环半实物联合仿真平台对DASS典型应用展开实验研究,充分验证了DASS的实时性、主动性与高效性.这对于降低交通拥堵、提升通行效率,加强车辆安全以及改善驾驶体验具有重要价值.
According to the lack of personalized, humanized and real-time problems of inadequate ＂request-response＂ current passive traffic services, the driving active service system（DASS） model of the hierarchical control ＂automatic recognition-active push＂ based on MAS（Multi-Agent system） is proposed. On the basis of modeling the driving environment entity and service, proposed the traffic service demand model based on planning graph matching problem, and researched on the algorithm of the key issues about service capabilities to match the users＇ needs. Through the independent construction of the driver in the loop Pre Scan-Logitech G27 semi physical joint simulation platform to collect data, using the traffic safety warning and real-time dynamic path planning experiment verified the real-time, active and efficient DASS. Research on DASS is conducive to reduce traffic congestion and improve traffic efficiency. At the same time, DASS can strengthen vehicle safety and improve the driving experience.
作者
梁军
赵振超
陈龙
徐永龙
朱宁
LIANG Jun ZHAO ZhenChao CHEN Long XU YongLong ZHU Ning(Automotive Engineering Research Institute, Jiangsu University, Zhenjiang 212013, China Department of Mechanical, Shizuoka Institute of Science and Technology, Fukuroi, Shizuoka 437-0032, Japan)
出处
《中国科学：技术科学》
EI
CSCD
北大核心
2016年第12期1303-1314,共12页
Scientia Sinica(Technologica)
基金
国家自然科学基金项目(批准号:U1564201
6157371
51108209)
交通运输部信息化项目(编号:2013-364-836-900)
中国博士后科学基金(编号:2016M600375)
江苏省高校优势学科建设工程项目(PAPD)
江苏省六大人才高峰项目(编号:DZXX-048)
全国统计科学研究项目(编号:2014596)
江苏省政府留学基金项目(编号:JS-2012-253)
江苏省自然科学基金项目(编号:BK20140570)资助
关键词
行车主动服务
MAS分层控制
车联网
driving active service
MAS hierarchical control
internet of vehicle
