多智能体协同高精地图构建关键技术研究  

Key technologies of multi-agent collaborative high definition map construction

作　　者：陈龙[1] 刘坤华 周宝定[2,3] 李清泉[4,5] CHEN Long;LIU Kunhua;ZHOU Baoding;LI Qingquan(School of Computer Science and Engineering, Sun Yat-sen Univercity,Guangzhou 510006, China;College of Civil and Transportation Engineering, Shenzhen University, Shenzhen 518060, China;Institute of Urban Smart Transportation & Safety Maintenance, Shenzhen University, Shenzhen 518060, China;Guangdong Key Laboratory of Urban Informatics, Shenzhen University, Shenzhen 518060, China;MNR Key Laboratory for Geo-Environmental Monitoring of Great Bay Area, Shenzhen 518060, China)

机构地区：[1]中山大学计算机学院,广东广州510006 [2]深圳大学土木与交通工程学院,广东深圳518060 [3]深圳大学城市智慧交通与安全运维研究院,广东深圳518060 [4]深圳大学广东省城市空间信息工程重点实验室,广东深圳518060 [5]自然资源部大湾区地理环境监测重点实验室,广东深圳518060

出　　处：《测绘学报》2021年第11期1447-1456,共10页Acta Geodaetica et Cartographica Sinica

基　　金：国家重点研发计划(2018YFB1305002);广东省自然科学杰出青年基金(2021B1515020020);国家自然科学基金(62006256,42171427);广州市重点研发项目(202007050002)。

摘　　要：自动驾驶车辆的自动化驾驶程度越高,对高精地图的要求越高。智能化的高精地图能够为L5级别自动驾驶车辆提供所需地图数据,是未来高精地图发展的重要方向。基于目前高精地图的构建方法,本文首先提出多智能体协同高精地图构建的定义,分析其构建框架。然后,对多智能体数据采集路径规划、多源异构一体化数据融合与表达、道路场景认知、智能高精地图融合、智能高精地图更新等关键技术进行了研究,提出了可行的技术方案。最后,分析了其未来构建过程中存在的挑战。The higher the automatic driving degree,the higher requirements of a high definition map.Intelligent high definition maps can provide information for L5 level autonomous vehicles,which is an important direction for the development of high definition maps in the future.According to the current high definition map construction methods,the definition of multi-agent collaborative intelligent high definition map construction is proposed.Its construction framework and key technologies:multi-agent routing for data collection,multi-source heterogeneous integration data fusion and expression,road scene cognition,intelligent high definition map fusion,intelligent high-definition update are studied,and their corresponding appropriate technical schemes are proposed.Besides,the challenges in the further construction process are analyzed.

关 键 词：高精地图 路径规划 道路场景理解 多源异构数据处理 

分 类 号：P208[天文地球—地图制图学与地理信息工程]