论文部分内容阅读
无人机组合导航滤波器的设计需要考虑器件和外部环境不稳定带来的影响,同时在飞行过程中也面临着组合导航系统噪声和量测噪声统计特性不确定问题,从而导致滤波精度低,稳定性差,甚至有可能发散,传统常规卡尔曼滤波无法解决上述问题。提出一种根据极大似然准则的自适应卡尔曼滤波算法,利用滤波残差的均值和方差不断对卡尔曼滤波的状态噪声方差阵和测量噪声方差阵进行实时修正,提高滤波器对模型不确定性和噪声变化的适应能力和鲁棒性。仿真表明,所提出的组合导航滤波器能够满足无人机导航任务的要求,并且具有很好的导航精度和稳定性。
The design of UAV integrated navigation filter needs to consider the impact of device and external environment instability, meanwhile it also faces the uncertainty of statistical characteristics of integrated navigation system noise and measurement noise during flight, resulting in low filtering accuracy, Poor stability, or even possible divergence, the conventional conventional Kalman filter can not solve the above problems. An adaptive Kalman filter algorithm based on maximum likelihood criterion is proposed. The state noise variance matrix of Kalman filter and the measurement noise variance matrix are continuously corrected by using the mean and variance of the residuals. Adaptability and robustness of deterministic and noise changes. The simulation shows that the proposed integrated navigation filter can meet the requirements of UAV navigation tasks and has good navigation accuracy and stability.