摘要
组合导航系统中,卡尔曼滤波算法的估计特性依赖于精确的系统模型和准确的外观测数据。任何一个条件不满足将导致滤波精度下降甚至发散。为此,本文引入进化神经网络和自适应卡尔曼滤波技术。仿真结果证明,文中所提算法能够有效克服常规卡尔曼滤波的缺点,并保持较高的精度。
In integrated navigation system, the performance of Kalman filter depends on the accurate system model and the proper observation data. Each absent will degrade the precision of conventional Kalman filter, and perhaps cause the filter divergent. So a kind of new adaptive Kalman filter based on evolutionary artificial neural networks was introduced in the system. Simulation results indicate that the algorithm proposed in this paper can efficiently overcome the shortcomings of conventional Kalman filter with a higher accuracy.
出处
《航天控制》
CSCD
北大核心
2008年第6期59-64,共6页
Aerospace Control
基金
国防973项目(973-61334)
国家自然科学基金资助项目(60374046)
国家自然科学基金项目(50575042)
关键词
组合导航系统
自适应滤波
进化规划
人工神经网络
Integrated navigation system （INS）
Adaptive filter
Evolutionary programming （ EP ）
Artificial neural networks （ANN）
