多模型估计理论

Multiple model estimation theory

研究利用多模型来逼近系统的动态性能,基于多模型设计组合导航系统的多模型自适应卡尔曼滤波器。通过仿真研究表明,在多模型自适应估计和交互式多模型滤波均克服了单一滤波由于外部扰动所造成的滤波误差过大甚至发散的问题。多模卡尔曼滤波器对组合导航系统能达到理想的控制精度、跟踪速度以及稳定性。研究表明,运用多模卡尔曼滤波能改善系统的瞬态响应,覆盖大范围的参数不确定性。

To support various applications, the multi-model Kalman filters for integrated inertial navigation system and multiply adaptive filter system are presented. Simulation illustrated that the algorithms multiple module estimation and interacting multiple model estimation can achieve stronger adaptability and higher accuracy than one of the single model in practice.