用于机载辅助导航的点云地形自适应分割方法研究  

An Adaptive Point-Cloud Terrain Segmentation Method for Auxiliary Airborne Navigation

作　　者：刘凯 张尉 武静[3] 沈晶晶[3] 李俊杰 LIU Kai;ZHANG Wei;WU Jing;SHEN Jingjing;LI Junjie(Science and Technology on Electro-Optic Control Laboratory,Luoyang 471023,China;Aviation Military Representative Office of Army Armament Department in Luoyang,Luoyang 471000,China;Luoyang Institute of Electro-Optical Equipment,AVIC,Luoyang 471000,China)

机构地区：[1]光电控制技术重点实验室,河南洛阳471023 [2]陆装驻洛阳地区航空军事代表室,河南洛阳471000 [3]中国航空工业集团公司洛阳电光设备研究所,河南洛阳471000

出　　处：《电光与控制》2021年第4期82-86,共5页Electronics Optics & Control

基　　金：陆航“十三五”预研项目(30104040205)。

摘　　要：针对目前机载辅助导航应用的激光雷达点云地形分割方法对不同地形自适应性差和实时性不高的问题,结合CSF算法和移动曲面拟合思想,提出了一种新的点云地形自适应分割方法。该方法通过格网划分和地形识别计算布料硬度值,使点云自适应分割;并利用分割过程中产生的布料模型生成用于头盔显示器显示的网格化地形。实验表明,与ISPRS公布的经典分割算法相比,该方法在一定准确率的基础上,具有地形自适应性强、实时性较高的优势,对激光雷达在辅助直升机导航方面的应用具有一定参考价值。In the field of auxiliary airborne navigationthe current Lidar-based point-cloud terrain segmentation method has low real-time performance and poor adaptability to different terrains.To solve the problemthis paper proposes an adaptive point-cloud terrain segmentation methodwhich combines the CSF algorithm with the fitting of moving surfaces.This method calculates the hardness value of the cloth through the gridding and recognition of terrains to adaptively segment the point cloudsand then uses the cloth model established in the segmentation process to generate terrain grids for helmet-mounted display.Experimental results show thatcompared with the classic terrain segmentation methods released by ISPRSthis method has the advantages of strong adaptability to different terrains and high real-time performance on the basis of ensuring certain accuracywhich has certain value in the application of the Lidar to auxiliary helicopter navigation.

关 键 词：机载辅助导航 激光雷达 地形点云分割 布料模拟 网格化地形 

分 类 号：V243.6[航空宇航科学与技术—飞行器设计] TP751.1[自动化与计算机技术—检测技术与自动化装置]