飞行器前视红外视觉导航基准图制备的理论与方法研究

项目名称： 飞行器前视红外视觉导航基准图制备的理论与方法研究

项目编号： No.61203189

项目类型： 青年科学基金项目

立项/批准年度： 2013

项目学科： 自动化学科

项目作者： 杨小冈

作者单位： 中国人民解放军第二炮兵工程大学

项目金额： 26万元

中文摘要： 前视红外视觉导航技术在飞行器的自主、智能导航领域体现出诱人的应用前景。项目以飞行器前视红外视觉导航为背景，从任务规划层面入手，研究导航区选取准则、导航基准图生成的方法与策略、导航基准图适配性评价等基础性问题。首先，依据飞行器导航任务性质，基于典型场景多源保障数据，如卫星影像、地形高程、三维模型等，进行典型场景信息融合及特性分析；其次，结合场景特性及飞行航迹约束条件，研究设计有效的准则参数，选定适合航空飞行的前视红外视觉导航区域；然后，针对导航区不同的场景或目标，研究利用多源数据生成高稳健性的模板、特征或知识基准图的方法；最后，结合飞行器动态特性及红外实时图特点，设计基准图适配性评价指标参数，进行多种模式下的导航仿真试验，对基准图的适配性进行评价。相关研究成果为飞行器前视红外视觉导航航迹规划及导航算法的研究奠定基础，对于推进前视红外视觉导航技术在飞行器中的应用具有重要意义。

中文关键词： 前视红外；视觉导航；选定准则；基准图；适配性分析

英文摘要： The Forward Looking Infrared (FLIR) vision-based navigation technology has a very attractive application prospect in the autonomous and intelligent navigation of aircraft. This project focuses on the mission planning of the aircraft FLIR vision-based navigation, and studies the selection role for the FLIR vision-based navigation area, the approach and strategy of the navigation reference image generation, and the suitability evaluation of navigation reference image, and so on fundamental problems. Firstly, according to the mission of aircraft navigation, the typical scene field information fusion and character analysis will be systemicly studied with the multi-source support data of typical scene field, such as satellite image, terrain elevation, and 3D model,etc.Secondly, considering the scene field characters and the flight route constraint conditions, the role parameters will be designed, which can be used for the aerial vehicle FLIR navigation areas selection.Furthermore, based on milti-source data, the novel algorithms will be proposed for high robust template, feature and knowledge refercenc image generation,which can deal with different target and scene field in navigation area.Finally, in term of aircraft dynamic characteristic and infrared real-time image character, reference image suitabiliy evaluation

英文关键词： FLIR；Vision-based Navigation；Selection Role；Reference Image；Suitability Analysis