基于平滑有界层的故障检测及抗差自适应滤波算法

Fault detection and robust adaptive filter algorithm based on smooth bounded layer

针对GNSS信号易受到干扰,卡尔曼滤波器性能降低、平滑变结构滤波算法在组合导航中应用受到原理限制的问题,提出了一种基于平滑有界层的故障检测及抗差自适应滤波算法。在对平滑变结构滤波算法研究分析的基础上,设计基于平滑有界层的量测信息质量评估指标,结合抗差估计思想构建等价权矩阵,在量测信息故障情况下,对组合导航系统进行重构和自适应调整,提高状态估计的精度和鲁棒性。仿真及实验验证结果表明:在GNSS各类典型故障下,所提故障检测算法均能够有效检测并快速响应,抗差自适应滤波算法在估计精度以及平稳性方面均优于基于残差卡方检测的自适应滤波算法,提高了SINS/GNSS组合导航系统的鲁棒性。在车载实验中,与基于残差卡方检测的自适应滤波算法相比,所提算法的定位精度提升约29%。

Aiming at the problems that GNSS signals are susceptible to interference,the performance of Kalman filter is degraded,and the application of smooth variable structure filter algorithm in integrated navigation is limited by principle,a fault detection and robust adaptive filter algorithm based on smooth bounded layer is proposed.Based on the analysis of the smooth variable structure filter algorithm,the measurement information quality evaluation index based on smooth bounded layer is designed and the equivalent weight matrix is constructed by combining robust estimation idea.In the case of measurement information failure,the integrated navigation system is reconstructed and adaptively adjusted to improve the accuracy and robustness of state estimation.Simulation and experimental verification results show that the proposed fault detection algorithm can effectively detect and respond quickly under various typical faults of GNSS.The robust adaptive filter algorithm is superior to the adaptive filter algorithm based on residual Chi-square detection in terms of estimation accuracy and stability,which improves the robustness of SINS/GNSS integrated navigation system.In the vehicle experiments,compared with the adaptive filter algorithm based on residual Chi-square detection,the positioning accuracy of the proposed algorithm is improved by about 29%.