飞行器地面试验运动参数视觉测量系统关键问题研究

发布时间：2018-05-07 17:51

本文选题：视觉测量 + 地面试验　；参考：《哈尔滨工业大学》2016年博士论文

【摘要】：飞行器地面试验是飞行器研制过程中不可或缺的一部分,通过飞行器地面试验可以有效提高飞行器的可靠性,并降低研制成本、缩短研制周期。在飞行器地面试验中,试验飞行器的位置、姿态等参数是运动学和动力学研究中的基本参数,可用于飞行器试验鉴定、事故分析以及工业设计等多个方面。视觉测量是目前广泛使用的一种运动参数测量方法。本文针对飞行器地面试验运动参数的高精度、大范围测量问题,在传统的运动参数视觉测量方法基础上,进一步探讨研究一种基于光束向量的运动参数测量方法,并对运动参数视觉测量系统研制过程中涉及的视觉测量系统的优化、大视场摄像机的标定、合作目标光束投影光斑的跟踪识别等问题进行研究。首先,对运动参数视觉测量系统的优化问题进行了研究。视觉测量系统的优化可以为系统的设计提供理论依据。在介绍了基于光束向量的飞行器地面试验运动参数测量原理及系统方案的基础上,充分考虑测量过程中的干扰因素,建立了以测量精度和测量范围为优化目标的测量系统多目标优化模型,并提出了一种基于多层次信息交互的多目标粒子群优化算法,给出了测量系统优化的Pareto最优解集,在规避干扰因素的同时,满足了系统不同测量精度和测量范围的设计需求,增强了系统的扩展性。其次,对大视场摄像机的现场标定问题进行了研究。为了满足飞行器地面试验运动参数视觉测量系统中大视场摄像机的高精度、快速、现场标定需求,分析了摄像机标定的相关理论基础,提出了一种基于平面单应性的大视场摄像机现场标定方法,该方法主要由基于平面单应性的大视场摄像机柔性标定与基于先验信息的标定参数现场校正两部分组成。上述方法仅通过靶标平面上少数特征点便实现了对大视场摄像机的现场标定,提高了标定效率,并且标定精度与传统摄像机标定方法相当,保证了标定精度。接着,对合作目标光束投影光斑的跟踪识别问题进行了研究。在测量过程中,为了获得合作目标光束的方向向量,需要对每条光束对应的投影光斑进行辨识,这一问题可通过对合作目标光束投影光斑跟踪识别来解决。在阐述了特征点跟踪识别问题以及传统的特征点跟踪识别方法(包括统计性方法和确定性方法)基础上,提出了一种基于多源信息的合作目标光束投影光斑跟踪识别方法。在跟踪识别过程中,该方法充分利用了光束投影光斑的位置预测信息、运动约束信息以及测量系统相关信息等多源信息,对光束投影光斑进出视场情况进行了有效处理,实现了合作目标光束投影光斑的准确跟踪识别。最后,对飞行器地面试验运动参数的大范围测量问题进行了研究。为了实现飞行器地面试验运动参数的高精度、大范围测量,在基于光束向量的运动参数测量方法基础上,增加合作目标以及合作目标中直线光束的数目,并提出了一种基于多合作目标的运动参数大范围测量方法。该方法有效地解决了运动参数大范围测量时弹体坐标系中合作目标光束方向向量的求取、参与运动参数解算的合作目标光束的选择以及旋转矩阵到欧拉角转换的奇异性等问题。同时,对运动参数测量过程中涉及的坐标系进行了分析,并给出了相应的坐标系标定及数据转换方法。最后,对本文飞行器地面试验运动参数视觉测量系统进行了测试,测试结果表明本文飞行器地面试验运动参数视觉测量系统能够实现对试验飞行器运动参数的高精度、大范围测量。
[Abstract]:The ground test of aircraft is an indispensable part of the development process of the aircraft. The ground test of the aircraft can effectively improve the reliability of the aircraft, reduce the development cost and shorten the development cycle. In the ground test of the aircraft, the parameters of the aircraft's position and attitude are the basic parameters in the research of kinematics and dynamics. It can be used in many aspects, such as aircraft test identification, accident analysis and industrial design. Visual measurement is a widely used measurement method of motion parameters at present. This paper aims at the high precision and large range measurement of the motion parameters of the aircraft ground test. On the basis of the traditional visual measurement method of the motion parameters, the research is further studied and studied. A method of measuring the motion parameters based on the beam vector is studied, and the optimization of the visual measurement system involved in the development of the visual measurement system of the motion parameters, the calibration of the large field of view camera and the tracking and recognition of the beam projection of the cooperative target are studied. First, the optimization of the visual measurement system of the motion parameters is carried out. The optimization of the visual measurement system can provide a theoretical basis for the design of the system. On the basis of introducing the principle and system scheme of the measurement of the motion parameters of the aircraft ground test based on the beam vector, the interference factors in the measurement process are fully considered, and the measurement system with the measurement precision and the measurement range as the optimal target is set up. A multi-objective particle swarm optimization algorithm based on multi-layer information interaction is proposed, and a Pareto optimal solution set for the optimization of the measurement system is given. At the same time, it meets the design requirements of different measurement precision and measurement range of the system, and increases the expansibility of the system. Secondly, the large field video camera is improved. In order to meet the high accuracy, rapid and field calibration requirements of the large field video camera in the visual measurement system of the aircraft ground test motion parameters visual measurement system, the theoretical basis of the camera calibration is analyzed, and a field calibration method for large field video camera based on plane single stress is proposed, which is the main method. It is composed of two parts, which are flexible calibration of large field camera based on plane homography and calibration parameter field correction based on prior information. The above method realizes the field calibration of large field video camera by only a few feature points on the target plane, and improves the calibration efficiency, and the calibration accuracy is equivalent to the traditional camera calibration method. The calibration accuracy is guaranteed. Then, the tracking recognition problem of the beam projection of the cooperative target is studied. In the process of measuring the direction vector of the target beam, it is necessary to identify the projection spot corresponding to each beam. This problem can be solved by the tracking recognition of the target beam projection light spot. On the basis of the traditional feature point tracking recognition and traditional feature point tracking identification (including statistical method and deterministic method), a multi source information based spot tracking recognition method for target beam projection is proposed. In the process of tracking recognition, this method makes full use of the position prediction letter of the beam projection spot. The multi source information such as interest, motion constraint information and the related information of the measurement system have been effectively processed, and the accurate tracking and recognition of the beam projection spot of the cooperative target is realized. Finally, the large range measurement of the motion parameters of the aircraft ground test is studied. On the basis of the measurement method of motion parameters based on the beam vector, the number of linear beams of the cooperative target and the cooperative target are increased on the basis of the method of measuring the motion parameters based on the beam vector. A large range measurement method based on the multi cooperative target is proposed. This method effectively solves the large range measurement of the motion parameters. The direction vector of the cooperative target beam in the projectile coordinate system is obtained, the selection of the cooperative target beam and the singularity of the rotation matrix to the Euler angle transformation are involved. Meanwhile, the coordinate system involved in the measurement of the motion parameters is analyzed, and the corresponding coordinate calibration and data conversion side are given. In the end, the visual measurement system of the ground test motion parameters of the aircraft is tested. The test results show that the visual measurement system of the ground test motion parameters of the aircraft can achieve high precision and wide range measurement for the motion parameters of the test aircraft.

【学位授予单位】：哈尔滨工业大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：V216.8;TP391.41

【相似文献】