Solar panel flexibility plays an important role in the attitude control of satellites. Therefore, traditionally the deformations of flexible solar panels are measured with a series of sensors along the panels itself. This paper presents a novel maiden attempt to simultaneously estimate the attitude and deformation of a flexible satellite using only 2 low-cost attitude sensors namely the sun sensor and magnetometer measurements. The flexible satellite is considered as a central rigid body with two attached flexible panels in order to derive the governing dynamic equations based on the Lagrange’s equation. Both Extended Kalman filter (EKF) and Unscented Kalman Filter (UKF) are employed for the purpose of simultaneous attitude and deformation estimations. By performing a comparison between EKF and UKF through a Monte-Carlo simulation, UKF shows less sensitivity during eclipse. However, its computational time is much more than that of EKF. The numerical results also demonstrate that including the flexibility model in the developed estimation equations enables the algorithm to successfully estimate the deformation at any solar panel location without the need for any conventional sensor that directly measures the strain and/or deformation.

太阳能电池板的灵活性在卫星的姿态控制中起着重要作用。因此，传统上，柔性太阳能电池板的变形是通过沿着电池板本身的一系列传感器来测量的。本文提出了一种新颖的首次尝试，即仅使用 2 个低成本的姿态传感器，即太阳传感器和磁力计测量，同时估计柔性卫星的姿态和变形。柔性卫星被认为是一个带有两个附加柔性面板的中央刚体，以便根据拉格朗日方程推导出控制动力学方程。扩展卡尔曼滤波器 (EKF) 和无迹卡尔曼滤波器 (UKF) 都用于同时进行姿态和变形估计。通过蒙特卡罗模拟对 EKF 和 UKF 进行比较，UKF 在日食期间表现出较低的敏感性。然而，它的计算时间比 EKF 多得多。数值结果还表明，在开发的估计方程中包含柔性模型使算法能够成功地估计任何太阳能电池板位置的变形，而无需任何直接测量应变和/或变形的传统传感器。