论文部分内容阅读
针对航天器姿态确定系统中存在较大初始误差及非线性较强的问题,提出了一种基于改进的正则化辅助粒子滤波(IRAPF)算法的航天器姿态确定方法。该算法将正则粒子滤波(RPF)与辅助粒子滤波(APF)相结合,将快速高斯变换(FGT)方法引入其中以减少计算量提高滤波的收敛速度。算法不仅有效地抑制了粒子退化问题,而且利用最新观测粒子来优化采样,并且在引入FGT后计算量与正则化的辅助粒子滤波相比降低了30%,改善了滤波的实时性。仿真结果表明了滤波的有效性。
Aiming at the problem of large initial error and strong nonlinearity in spacecraft attitude determination system, a spacecraft attitude determination method based on improved Regularized Aided Particle Filter (IRAPF) algorithm is proposed. This algorithm combines regular particle filter (RPF) with assistant particle filter (APF), and introduces a fast Gaussian transform (FGT) method to reduce the computational complexity and improve the convergence speed of the filter. The algorithm not only effectively suppresses the problem of particle degeneration, but also optimizes the sampling by using the latest observed particles. Compared with the regularized Auxiliary Particle Filter, the computational cost is reduced by 30% after FGT is introduced, which improves the real-time performance of the filter. The simulation results show the effectiveness of the filtering.