摘要
智能化的无人系统在现代社会中起着重要作用,而对环境信息的准确感知以及自身位置的精准估计是无人系统智能、高效执行任务的核心基础,视觉与激光雷达传感器是无人系统常用的感知与导航传感器。近年来,随着应用场景的拓展,无卫星信号、无光等恶劣环境对无人系统的感知与自主导航技术提出了新的挑战。针对上述环境,对无人系统中视觉/激光雷达感知技术与自主导航技术及其进展进行了分析和总结。从感知和导航定位两个应用层面出发,深入讨论和分析了基于视觉和激光雷达手段的深度估计、目标检测、自主导航、地图构建等技术的机理差异、实现方法及特性。分析了目前国内外研究成果及进展,总结对比了目前的技术特点与局限性,并展望了未来无人系统感知与导航的关键技术挑战与发展趋势。
The intelligent unmanned system plays an important role in modern society.Accurate perception of environmental information and accurate position self-estimation are the core foundation of the unmanned system for its intelligence and high efficiency.Visual and LiDAR sensors are common perception and navigation sensors for unmanned systems.In recent years,with the expansion of application scenarios,GNSS-denied,no light,and other hostile environments pose new challenges to the perception and navigation technology of unmanned systems.To tackle these challenges,the visual/LiDAR perception technology and autonomous navigation technology and their progress in unmanned systems are analyzed and summarized.From two application aspects of perception and navigation,the mechanism differences,implementation methods and characteristics of depth estimation,target detection,autonomous navigation and mapping technologies based on vision and LiDAR are discussed and analyzed.The current research progress at home and abroad is analyzed.Besides,the current technical characteristics and limitations are summarized and compared,and the key technical challenges and development trends of unmanned system perception and navigation in the future are forecasted.
作者
赖际舟
袁诚
吕品
刘建业
何洪磊
LAI Ji-zhou;YUAN Cheng;LYU Pin;LIU Jian-ye;HE Hong-lei(College of Automation Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China)
出处
《导航定位与授时》
CSCD
2021年第3期1-14,共14页
Navigation Positioning and Timing
基金
国家自然科学基金(61973160)
航空科学基金(2018ZC52037)
工信部民机专项(2018-S-36)。
关键词
无人系统
不依赖卫星
环境感知
自主导航
同步定位与地图构建
Unmanned system
Independent of GNSS
Environmental perception
Autonomous navigation
Simultaneous localization and mapping(SLAM)
