基于动态攻击区优化计算的空空导弹导引制导一体化新机理探索

This project intends to focus on two problems that: 1) the midcourse guidance in air-to-air missile (AAM) is deep coupled with target motion tracking, and 2) the midcourse guidance parameters provided by airborne radar fire control system do not match with those actual target motion parameters. Firstly, a new mechanism of feedback and cooperation between state-estimation/mode-recognition of target, and midcourse guidance parameter optimization of AAM, will be explored and studied. To support the new mechanism, this project further concentrates on researching two kinds of foundational theory or key technology, i.e. that 1) joint state estimation and mode recognition of target motion, and 2) joint optimization of both dynamic attack area (DAZ) and midcourse guidance parameters for AAM. Specially, such the theory or technology is achieved by breaking through two corresponding difficulties, including higher-order moment fitting of equivalent unknown input caused by target motion mode uncertainty, and construction and optimization of DAZ's likelihood probability evolution function under multi-variables and multi-constraints. Thirdly, by doing above, the new coupling and interconnection mechanism among state estimation, motion mode recognition and midcourse parameter optimization, is revealed. In addition, with DAZ computation optimized, it can be realized that the intelligent planning of flight trajectory in AAM's midcourse guidance, and that the adaptive selection of intercept point of starting AAM's terminal guidance, for enhancing the decision-making ability of AAM to cope with increasingly severe situation change of both AAM and target. Finally, this project may be helpful to the cross integration of multi disciplines, such as estimation, identification and optimization, and to provide a new mechanism, a new concept and a new method for defense engineering application of airborne weapon system.

本项目拟围绕超视距空战下空空导弹（AAM）中继制导与目标运动导引存在深度耦合、机载火控系统提供的中继制导参数与目标实际运动参数不匹配等问题，探索目标状态估计、模式识别与AAM中继制导参数优化等信息环节之间反馈/协同的闭环一体化处理新机理，重点研究联合目标运动状态估计与模式识别、联合优化AAM动态攻击区（DAZ）与中继制导参数等相关基础理论与关键技术，突破目标运动模式不确定的等价未知干扰高阶矩拟合、多变量多约束下的DAZ似然概率演化函数构建与优化等难点，揭示状态估计、模式识别、参数优化等环节之间的耦合关联新机理，在优化计算DAZ的同时，实现AAM中继飞行弹道智能规划、末制导开机拦截点自适应选择，提升AAM应对超视距空战下弹目态势变化日趋剧烈的决策能力。助力于估计、辨识与优化等多学科的交叉融合，为机载武器系统攻防对抗的国防工程应用提供新机理、新概念与新方法。

空空导弹作为超视距空战中制夺取制空权的重要利器，是未来战争中决胜千里的核心武器装备。传统空空导弹制导方式面临目标导引与导弹制导深度耦合、制导参数与目标参数失配的重大挑战，不仅命中率低，且鲁棒性差。本项开展目标状态估计、干扰模式识别与空空导弹制导轨迹优化等前沿基础理论与算法研究，设计实时动态攻击区搜索框架，构建了联合目标状态估计与模式识别的理论迭代框架体系，实现了自适应模式识别大幅度提高了状态估计精度与抗干扰能力，以“自身导引+ 超前指令修正”的方式提高了导弹索敌能力；搭建了模块化和平台化的动态攻击区计算与空空导弹导引制导一体化验证系统，揭示一体化策略比传统割裂式策略在鲁棒性、精度与命中率方面的优越性。发表论文11篇，其中SCI源期刊7篇，中科院Top期刊4篇，Automatica长文1篇，在包括国际信息融合会议、国际控制与自动化会议等国际知名会议上发表论文4篇；计划出版学术专著1部，目前正在签订合同，计划于2023年12月31日前出版；申请国家发明专利18项，其中授权4项；培养硕士研究生11名，博士研究生6名；组织本领域国际国内学术会议5次，邀请国内外同行学者来课题组开展学术交流和合作研究4人次。