基于无人机多传感吊舱系统的火点实时定位方法  

Method of fire point real-time positioning based on multi-sensors system of UAV

作　　者：李兴东 肖旭 陈显东 张时雨 刘九庆[1,2] LI Xingdong;XIAO Xu;CHEN Xiandong;ZHANG Shiyu;LIU Jiuqing(School of mechanical and electrical engineering, Northeast Forestry University, Harbin 150006, China;Research Center of National Forestry Grassland Artificial Intelligence and Equipment Engineering Technology, Northeast Forestry University, Harbin 150006, China)

机构地区：[1]东北林业大学机电工程学院,哈尔滨150006 [2]国家林业草原人工智能与装备工程技术研究中心,哈尔滨150006

出　　处：《火灾科学》2022年第1期35-45,共11页Fire Safety Science

基　　金：黑龙江省自然科学基金(TD2020C001);中央高校基本科研业务费专项基金(2572019CP20)。

摘　　要：林火定位是林火智能监测设备的核心技术。提出了一种基于激光雷达、红外热像仪及组合惯导多数据融合的火点定位方法。首先设计了一个通用的无人机吊舱系统,并基于ROS框架实现数据采集、数据处理和数据传输等功能;其次提出了一个基于无人机吊舱系统的火点全局定位方案,根据红外热像仪成像特性识别火点,同时将密集点云与红外图像进行数据融合,估计出火点的三维空间位置;然后根据无人机位姿提出了一种基于墨卡托投影的火点全局定位方案,得到了火点的GPS位置。最后通过试验得到了该方案的全局定位精度:在实验中经度最大误差为2.36×10^(-5)°,纬度最大误差为1.84×10^(-5)°,高程最大误差为0.9261 m,为其他林火定位方案提供了技术支持。The forest fire location is the core technology of forest fire intelligent monitoring equipment.This paper presents a fire point location method based on multi-data fusion of lidar,infrared thermal imager and integrated inertial navigation system.Firstly,this paper designs a general UAV system which realizes the functions of data acquisition,data processing and data transmission based on ROS framework.Secondly,a fire point global location scheme based on UAV system is proposed.The fire point is identified according to the imaging characteristics of infrared thermal imager.At the same time,the dense point cloud and infrared image are fused to estimate the three-dimensional spatial position of the fire point.Then,according to the pose of UAV,a global positioning scheme of fire point based on Mercator projection is proposed,and the GPS position of fire point is obtained.Finally,the global positioning accuracy of the scheme is obtained through experiments:the maximum error of longitude is 2.36×10^(-5)°,the maximum error of latitude is 1.84×10^(-5)°,and the maximum error of elevation is 0.9261 m,which provides technical support for other forest fire positioning schemes.

关 键 词：多传感器方法 数据融合 林火监测 无人机吊舱 林火全局定位 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]