基于低空遥感的丘陵山地田间道路路径规划研究  

A Study on path planning for field roads in hilly and mountainous areas based on low-altitude remote sensing

作　　者：柴昕龙 朱建锡 么炜[1] 陈长卿 CHAI Xinlong;ZHU Jianxi;YAO Wei;CHEN Changqing(College of Information Science and Technology,Agricultural University fo Hebei,Baoding 071001,China;Zhejiang Academic of Agricultural Machinery,Jinhua Zhejiang 321013,China;Jinhua Key Laboratory of Applicable Agricultural Machinery for Hilly Mountainous Areas,Jinhua Zhejiang 321013,China)

机构地区：[1]河北农业大学信息科学与技术学院,河北保定071001 [2]浙江省农业机械研究院,浙江金华321017 [3]金华市丘陵山地适用农机重点实验室,浙江金华321017

出　　处：《河北农业大学学报》2025年第1期97-106,共10页Journal of Hebei Agricultural University

基　　金：浙江省“领雁”研发攻关计划(2023C02049).

摘　　要：路径规划是农用车辆自主导航的基础,现有的路径规划方法局限于二维平面,忽略地形因素对导航任务的影响。研究旨在通过构建丘陵山地倾斜摄影模型,将低空遥感技术与农用车辆的路径规划任务相结合。主要包括三个步骤,首先,通过无人机对行驶区域进行倾斜摄影构建三维实景模型,获取行驶区域的真实数字正射影像和数字高程模型。其次,通过图像处理的方法获取道路环境信息包括道路走向、道路边界等。最终,通过改进RRT*算法进行路径规划,从而得到行驶路线,并将三维信息附加在路径上,生成三维路径。实验表明,所提出的方法可以适用于丘陵山地田间道路的直线路段,及多曲率复杂路段。依靠无人机获取数据,无人车搭载的GNSS接收器进行定位,减少了对多传感器的依赖性,形成了全局规划策略。三维路径提供的高程信息能让农用车辆全面了解自身位置和目的地位置。本文成果可为丘陵山地田间道路自主导航提供新思路。Path planning is the basis of autonomous navigation for agricultural vehicles,and existing path planning methods are limited to a two-dimensional plane ignoring the influence of terrain factors on the navigation task.The study aims to combine low-altitude remote sensing technology with the path planning task of agricultural vehicles by constructing an inclined photography model of hilly mountainous terrain.It mainly consisted of three steps.Firstly,a three-dimensional realistic model was constructed by tilt photography of the driving area by UAV to obtain the real digital orthophoto and digital elevation model of the driving area.Secondly,the road environment information including road direction and road boundary was obtained by image processing method.Finally,the path planning was performed by designing the improved RRT*algorithm to obtain the traveling route,following attachment of the 3D information of the path.Experiments showed that the proposed method can be applied to straight sections of field roads in hilly mountainous areas and complex sections with multiple curvatures.Relying on the UAV to acquire data and the GNSS receiver carried by the unmanned vehicle for localization reduced the dependence on multiple sensors and formed a global planning strategy.The elevation information provided by the 3D path enables the agricultural vehicle to have a comprehensive understanding of its position and destination location.The results of this paper can provides new ideas for autonomous navigation on field roads in hilly and mountainous areas.

关 键 词：丘陵山地 自主导航 精准农业 3D路径规划 倾斜摄影 

分 类 号：S24[农业科学—农业电气化与自动化]