临近空间多拦截器协同探测问题研究

发布时间：2018-01-07 10:13

本文关键词：临近空间多拦截器协同探测问题研究　出处：《哈尔滨工业大学》2015年硕士论文　论文类型：学位论文

【摘要】：在可以预见的未来里,由于科技水平不断发展使武器具有更佳的性能,未来将迎来愈发复杂的空战环境。而在临近空间内,高超声速飞行器具有目标速度与响应时间快、目标探测困难以及目标的侧向机动能力强等特点,是对国家安全有巨大威胁的一种武器。因此在研究临近空间高超声速飞行器的同时亦有必要研究其防御技术。综合目前精确制导武器的技术现状、临近空间高超声速飞行器的拦截难点表明,目前精确制导武器很难实现对高超声速目标一对一的直接碰撞拦截。当一对一拦截时,由于单个拦截器对目标探测精度较低,导致拦截概率低;而多个拦截器协同探测可以提高对相对运动状态的探测精度,进而提高对目标的拦截概率。因此,在这种背景下研究多飞行器协同探测以提高对临近空间高超声速目标的探测精度具有重要的现实意义。本文主要完成了以下三方面的内容。首先,本文基于几何关系对多拦截器协同制导问题进行了描述并建立起独立运动模型与相对运动模型。考虑到实际可测量的数据并兼顾滤波简单,选择状态变量、量测量以及协同变量使滤波精度最高。针对飞行器独立运动选择全局惯性坐标系建立起拦截器运动模型以及目标运动模型,针对弹-目相对运动分别选择视线坐标系以及惯性坐标系建立起弹-目相对运动的数学模型。其次,研究基于交互多模型的滤波算法,拦截器之间进行数据共享以便协同完成对目标区域的探测。由于协同探测融合了多个拦截器的探测信息,因而能够增大拦截器的探测范围并提高对目标状态探测的精度。通过交互式多模型算法进行滤波估计后,能够获得弹-目相对信息的估计值,为逼近目标的真实运动轨迹打下基础。同时,采用简单模型通过概率组合来逼近目标真实运动,研究如何在采用较少模型条件下获得较高精度,并通过仿真验证所选择的滤波所用模型能否在保证实时性的同时,最好地近似临近空间高超声速飞行器的实际飞行轨迹。最后,通过数据融合算法提高对相对运动状态的估计精度。将不同拦截器的探测信息进行数据融合,由于滤波过程中每一个拦截器与目标间皆可通过滤波获得相对运动状态的估计,因此可以以相对运动状态估计协方差矩阵大小作为加权依据,对所获得的多个数据进行融合,以得到更高精度的结果,并通过仿真验证方法的有效性。
[Abstract]:In the foreseeable future, due to the continuous development of science and technology to make weapons with better performance, the future will usher in increasingly complex air combat environment, and in the near space. Hypersonic vehicle has the characteristics of fast target speed and response time, difficult target detection and strong lateral maneuverability. Therefore, it is necessary to study the defense technology of hypersonic vehicles in the near space at the same time, and to synthesize the present technical situation of precision guided weapons. The difficulty of intercepting near space hypersonic vehicle shows that it is very difficult to achieve direct collision interception of hypersonic target with precision guided weapon at present, when intercepting one to one. Because of the low detection accuracy of single interceptor, the interception probability is low. The cooperative detection of multiple interceptors can improve the detection accuracy of the relative moving state, and then improve the probability of intercepting the target. In this context, the study of multi-vehicle cooperative detection to improve the detection accuracy of hypersonic targets near the space has important practical significance. This paper mainly completes the following three aspects. In this paper, the cooperative guidance problem of multiple interceptors is described based on geometric relations, and an independent motion model and a relative motion model are established. Considering the actual measurable data and taking into account the simple filtering, the state variables are selected. The precision of filtering is the highest due to the measurement and cooperative variables. The interceptor motion model and the target motion model are established by selecting the global inertial coordinate system for the independent motion of the vehicle. Aiming at the relative motion of projectile and object, the line of sight coordinate system and the inertial coordinate system are selected to establish the mathematical model of the relative motion. Secondly, the filtering algorithm based on interactive multi-model is studied. Data sharing between interceptors in order to cooperate to complete the detection of the target region. The cooperative detection fusion of the detection information of multiple interceptors. Therefore, the detection range of interceptor can be increased and the precision of target state detection can be improved. After filtering estimation by interactive multi-model algorithm, the estimated value of projectile relative information can be obtained. At the same time, the simple model is used to approximate the real motion of the target through the combination of probability, and how to obtain higher accuracy under the condition of using fewer models is studied. And through the simulation to verify whether the selected filtering model can guarantee real-time, and best approximate to the actual flight trajectory of the space hypersonic vehicle. Finally. The data fusion algorithm is used to improve the estimation accuracy of relative motion state. The detection information of different interceptors is fused. Because the estimation of relative motion state can be obtained by filtering between each interceptor and target in the process of filtering, the size of covariance matrix of relative motion state estimation can be used as the weighting basis. The multiple data are fused to obtain higher accuracy results, and the effectiveness of the method is verified by simulation.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：V249;E926

【参考文献】