基于SRUKF的偏心仅测角相对导航方法

发布时间：2018-05-05 10:49

本文选题：仅测角导航 + 自主交会　；参考：《宇航学报》2016年11期

【摘要】：针对以单个光学/红外相机作为相对导航敏感器的航天器,提出采用相机偏心安装和平方根无损卡尔曼滤波(SRUKF)两种方法,解决自主交会时中远距离和近距离全过程的仅测角导航(AON)问题。在观测相机安装时,人为设置一个分米量级的安装偏心,按照观测距离分段建立相机观测模型并推导了观测敏感矩阵,从理论上证明了径向和法向相机安装偏心可以改善近距离v-bar相对位置保持点的观测度,并且对远距离导航精度没影响。构建了基于SRUKF的仅测角导航方法,该方法不需要强制要求初始估计误差满足小量假设和零均值高斯分布假设。经算例验证,在远距离时,采用本文提出的仅测角导航方法,其导航精度比传统扩展卡尔曼滤波(EKF)方法提高了一倍,计算负载降低了约10%;在近距离时,解决了仅测角导航v-bar相对位置保持点不可观测问题;并且当初始估计误差呈非高斯特性时,依然可以实现对相对距离的无偏估计。
[Abstract]:For a spacecraft with a single optical / infrared camera as a relative navigation sensor, two methods, the eccentric installation of the camera and the square root lossless Kalman filter (SRUKF), are proposed. To solve the problem of angle-only navigation in the whole process of long distance and close range in autonomous rendezvous. In the installation of the observation camera, an artificial installation eccentricity of the order of decimeter is set up, the observation model of the camera is established according to the observed distance segment and the observation sensitivity matrix is derived. It is proved theoretically that the eccentric installation of radial and normal cameras can improve the observability of the relative position holding point of v-bar in close range, and has no effect on the accuracy of long range navigation. An angle-only navigation method based on SRUKF is proposed. The method does not require the initial estimation error to meet the small assumption and the Gao Si distribution hypothesis of zero mean value. The numerical examples show that the accuracy of the proposed angle-only navigation method is twice as high as that of the traditional extended Kalman filter (EKF) method, and the computational load is reduced by about 10 parts at close range. The problem of unobservable relative position holding point of v-bar for angle-only navigation is solved, and the unbiased estimation of relative distance can still be realized when the initial estimation error is non-.
【作者单位】：国防科学技术大学航天科学与工程学院;瑞士联邦理工学院瑞士宇航中心;载人航天总体研究论证中心;
【基金】：国家自然科学基金(11572345,11272346) 国家部委资助项目(2013CB733100)
【分类号】：V448.2

【参考文献】