首页 | 本学科首页   官方微博 | 高级检索  
     检索      

面向智能驾驶的高精度多源融合定位综述
引用本文:许智理,闫倬豪,李星星,申志恒,周宇轩,吴宗洲,李昕.面向智能驾驶的高精度多源融合定位综述[J].导航定位于授时,2023,10(3):1-20.
作者姓名:许智理  闫倬豪  李星星  申志恒  周宇轩  吴宗洲  李昕
作者单位:武汉大学测绘学院,武汉 430079
基金项目:国家重点研发项目(2021YFB2501100);湖北珞珈实验室专项基金资助项目(220100006)
摘    要:随着信息时代各行各业效率的提升,传统的人工驾驶交通系统已逐渐无法满足人们对高效率、低风险交通服务的需求,而智能驾驶技术的出现为这一领域带来了机遇。如今,以自动驾驶为代表的智能驾驶已经成为一种实用的深度交叉技术,其核心模块包括高精度定位、场景感知、决策规划与控制等。定位模块作为智能驾驶系统中最基本、最核心的功能模块,需具备高精度、高可用、低时延的性能特点。当前,结合高精度卫星导航、惯性导航以及环境感知的多源融合技术已成为实现泛在智能驾驶所公认的核心手段,通过充分利用车载传感器的量测信息可以实现精确、可靠的定位服务。从导航定位中常用的传感器技术出发,对当前智能驾驶领域涉及的高精度定位技术进行了全面的回顾,给出了主流的基于滤波和因子图优化的多源融合框架,并对代表性算法进行了整理。最后,总结了现阶段智能驾驶中高精度定位技术的发展现状,并对未来的发展趋势进行了展望。

关 键 词:多源融合  智能驾驶  全球卫星导航系统  惯性导航  同步定位与地图构建

Review of high precision multi-sensor integrated positioning towards intelligent driving
XU Zhili,YAN Zhuohao,LI Xingxing,SHEN Zhiheng,ZHOU Yuxuan,WU Zongzhou,LI Xin.Review of high precision multi-sensor integrated positioning towards intelligent driving[J].Navigation Positioning & Timing,2023,10(3):1-20.
Authors:XU Zhili  YAN Zhuohao  LI Xingxing  SHEN Zhiheng  ZHOU Yuxuan  WU Zongzhou  LI Xin
Institution:School of Geodesy and Geomatics, Wuhan University, Wuhan 430079, China
Abstract:With the efficiency improvement of industries in the information era, the traditional human-driven traffic system has gradually failed to meet people''s demand for high-efficiency and low-risk traffic services, and the emerging intelligent driving technology has brought opportunities to this field. Nowadays, intelligent driving, represented by autonomous driving, has become a practical and deep crossing technology. Its core modules include high-precision positioning, scene perception, planning, control and other aspects. As the most basic and core functional module in an intelligent driving system, the positioning module needs to be of high-precision, high-availability, and low-latency performance. At present, the multi-sensor fusion technology integrated with high-precision satellite navigation, inertial navigation and environmental perception has become the recognized core means to realize ubiquitous intelligent driving, and accurate and reliable positioning services can be achieved by making full use of the measurement information of on-board sensors. Starting from the sensor technology commonly used in navigation and positioning, we comprehensively review the high-precision positioning technology involved in the current field of intelligent driving, present the mainstream multi-sensor fusion framework based on filtering and factor graph optimization and sort out representative algorithms. Finally, we summarize the current development status of high-precision positioning technology in intelligent driving and prospect the development trend in the future.
Keywords:
点击此处可从《导航定位于授时》浏览原始摘要信息
点击此处可从《导航定位于授时》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号