基于视觉信息的航天器位姿估计迭代算法

A Novel Binocular OI(Orthogonal Iteration) Fusion Algorithm Better for Estimation of Spacecraft Position and Attitude

基于视觉信息的空间目标相对位姿估计问题是未来空间操作的关键所在,正交迭代算法是一种具有实时性好且全局收敛特点的单目位姿估计算法。为了有效利用多个摄像机获取的数据,进一步提高位姿估计算法的综合性能,提出了一种双目正交迭代融合算法,取两个摄像机获取的所有特征点的目标空间共线性误差平方和作为误差函数,推导得到使两个摄像机总的目标空间共线性误差最小的迭代求解过程,进而采用单位四元数法求解算法中的目标函数得到目标位姿的估计值,从根本上解决传统算法旋转矩阵计算误差大的问题。最后通过仿真算例验证了该算法在计算精度、抗噪性能以及稳定性等方面具有的优势。

The introduction of the full paper reviews some papers in the open literature and then proposes a novel OI fusion algorithm, which we believe is better. Sections 1 and 2 explain our better algorithm. Section 1 briefs OI algorithm. The core of section 2 consists of： （ 1 ） we use as error function, which is given by eq. （ 10）, the quadratic sum of collinearity errors of all the feature points in target space obtained from two cameras; （2） we deduce the four-step iterative procedure to get the minimum target space collinearity error as indicated in eq. （12） ; （3） we estimate the target position and attitude through unit quaternion of the target, thus avoiding fundamentally the troublesome rotation matrix error in traditional algorithms. Simulation results, presented in Figs. 2 through 5 and Table 1, and their analysis demonstrate preliminarily that our novel OI fusion algorithm is significantly better in precision, anti-noise ability and stability performances.