摘要
考虑到组合导航系统的噪声具有非先验性,而传统的卡尔曼滤波器要求假设动态模型和观测模型的噪声统计特性已知,提出了用前向神经网络来辅助调节卡尔曼滤波器,使其具有自适应能力以应付动态环境的扰动。仿真研究表明:该算法优于标准卡尔曼滤波器。
The problem of integrated navigation system with uncertain noise is considered. The conventional Kalman filter assumes that the statistical properties of the noise in dynamic model and observation system are exactly known. The paper presents a method, which uses neural network-aided Kalman filtering scheme for the adaptive capability to deal with the disturbance in dynamic situation. The simulation study shows that the arithmetic is better than that of standard Kalman filtering.
出处
《中国惯性技术学报》
EI
CSCD
2003年第2期40-43,共4页
Journal of Chinese Inertial Technology