农业无人机动态降落方法研究  

Dynamic Landing Method of Agricultural UAV

作　　者：万晓燕 石永康 赵玉花 李鹏[1] Wan Xiaoyan;Shi Yongkang;Zhao Yuhua;Li Peng(School of Mechanical Engineering,Xinjiang University,U··r u··mqi 830046,China)

机构地区：[1]新疆大学机械工程学院,乌鲁木齐830046

出　　处：《农机化研究》2024年第7期21-26,共6页Journal of Agricultural Mechanization Research

基　　金：国家自然科学基金项目(51965056);自治区高层次人才项目(100400027);自治区高校科研计划项目(61021800081);新疆大学科研启动项目(620312351)。

摘　　要：为实现农业无人机高速、准确降落于地面移动机械设备(自走式施肥机构、农业无人车等)上,基于Aruco码设计了一种复合视觉识别标志,并基于该视觉识别标志提出了农业无人机动态降落控制策略。降落策略主要分为两个阶段:阶段一为标识识别阶段,使用Aruco标识检测算法进行标识识别;阶段二为对带标识地面移动机械设备的稳定跟踪。工作时,利用PnP类算法求解无人机当前位姿,使用PD轨迹追踪算法实现农业无人机对地面移动平台的追踪。为了验证该控制策略的有效性与准确性,利用开源仿真平台Prometheus进行仿真验证,并使用MatLab进行数据分析,结果表明:该控制策略可实现农业无人机的自主降落。In order to achieve high-speed and accurate landing of agricultural drones on ground mobile mechanical e-quipment(self-propelled fertilization mechanisms,agricultural unmanned vehicles,etc.).Based on the Aruco code,this paper designs a composite visual identification mark,and based on the visual identification mark,a dynamic landing con-trol strategy for agricultural drones is proposed.The landing strategy is mainly divided into two stages:the first stage is the identification identification stage,which uses the Aruco identification detection algorithm to identify the identification;the second stage is the stable tracking of the ground mobile mechanical equipment with the identification.The PnP algorithm is used to solve the current pose of the UAV,and the PD trajectory tracking algorithm is used to realize the tracking of the ground mobile platform by the agricultural UAV.In order to verify the effectiveness and accuracy of the control strategy,the open source simulation platform prometheus is used for simulation verification.Using Matlab for data analysis,the data results show that the control strategy can realize the autonomous landing of agricultural drones.

关 键 词：农业无人机 动态自主降落 位姿估计算法 PID轨迹跟踪 

分 类 号：S252[农业科学—农业机械化工程]