自主多旋翼无人机系统:感知、规划与控制  

Autonomous multirotor UAV systems:perception,planning and control

作　　者：郭晶晶 刘允刚[1] 满永超[1] 陈琳 GUO Jing-jing;LIU Yun-gangy;MAN Yong-chao;CHEN Lin(School of Control Science and Engineering,Shandong University,Jinan Shandong 250061,China)

机构地区：[1]山东大学控制科学与工程学院,山东济南250061

出　　处：《控制理论与应用》2024年第10期1707-1725,共19页Control Theory & Applications

基　　金：国家自然科学基金项目(62033007,61821004,62373218);山东省重大基础研究项目(ZR2023ZD37);中国高校产学研创新基金项目(2021ZYA02007)资助.

摘　　要：无人机成本低、机动性强且视野广阔,广泛用于军事和民用领域.不断增强感知与决策能力,进而提高无人机的自主性,是无人机系统发展的必然趋势.本文围绕实现多旋翼无人机自主控制所面临的关键科学技术问题,即灵敏稳健的飞行控制、实时精准的感知与估计和高效的在线运动规划,展开全面概述.从硬件和软件系统,到飞行控制、感知与估计、运动规划的具体算法,试图构建多旋翼无人机自主控制系统的整体认知体系,旨在为相关理论的研究提供系统性的见解,并为多旋翼无人机自主控制系统的设计提供技术参考.最后,总结了现阶段多旋翼无人机自主控制技术的局限性,并对未来发展趋势进行了展望.Unmanned aerial vehicles(UAVs)have been ubiquitous in military and civilian fields owing to their low cost,high maneuverability and wide visibility.As an inevitable development trend,UAVs should increasingly enhance perception and decision-making abilities,and in turn improve their autonomy.This paper presents a comprehensive overview of autonomous control for multirotor UAVs centring on key scientific and technical issues,namely,sensitive and robust flight control,real-time accurate perception and estimation,and efficient online motion planning.The attempt is to construct an overall cognitive architecture of autonomous control systems for multirotor UAVs,from hardware and software systems to specific algorithms of flight control,perception and estimation,and motion planning.The aim is to provide systematic insights for related theoretical research and technical reference for autonomous control system design of multirotor UAVs.Finally,the current limitations of autonomous control techniques of multirotor UAVs are summarized,and future development trend is prospected.

关 键 词：无人机 自主控制 感知与估计 飞行控制 运动规划 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置] V279[自动化与计算机技术—控制科学与工程] V249[航空宇航科学与技术—飞行器设计]