自适应交互多模型滤波在INS/CNS组合导航中的应用

Adaptive interactive multiple model and its application in INS/CNS integrated navigation system

针对惯性/天文(INS/CNS)组合导航系统中由于量测噪声的统计特性随应用环境发生变化而导致滤波精度降低的的问题,提出了一种基于自适应交互多模型(AIMM)的组合导航算法,该算法将改进的Sage-Husa自适应滤波器与交互多模型相结合,以少量的模型实现对实际模态的覆盖。仿真结果表明:与常规Kalman滤波方法相比,该方法较好地覆盖了实际模态变化情况,实现了对量测噪声统计特性变化的自适应调整,在观测值发生异常时,导航参数误差没有大的跳变,导航滤波器在观测值异常时具有一定的自适应能力,从而保证了导航滤波器的可靠运行,提高了导航系统的精度和可靠性。

According to the problem that the measured noise statistical characteristics differ with the change of application enviroment in inertial/celestial（INS/CNS） navigation system and the filter accuracy was decreased,an adaptive interactive multiple model（AIMM） algorithm was presented.The AIMM algorithm which combined the IMM algorithm with the improved Sage-Husa adaptive filtering algorithm can achieve the coverage of real situation well through few sub-models.The simulation results demonstrate that the algorithm can cover the change of the real situation better than the conventional Kalman filter.Stronger adaptive ability of navigation filter can be obtained by the adaptive algorithm when the observation was abnormal.The errors of navigation parameters have no significant change when observed noise statistical characteristics changed singularly.The navigation filter has a certain adaptive ability when the observations change.So the algorithm can ensure that the navigation filter operates reliably,and the accuracy and reliability of integrated navigation system are improved as well.