基于STF-Singer模型的AUV传感器故障诊断

AUV sensor fault diagnosis based on STF-Singer model

为了实现水下机器人传感器系统的故障诊断和信号恢复,将基于信号处理的方法和强跟踪滤波器方法相结合,提出了一种基于一阶时间相关函数的辛格模型和强跟踪滤波器理论的STF-Singer模型的故障诊断方法,该方法不依赖AUV精确的数学模型,对于AUV传感器故障有着很好的辨识能力。某型水下机器人的计算机仿真和海上试验的数据结果,验证了该方法的有效性和可行性。

To deal with the fault diagnosis and signal recovery issue of the sensor system for autonomous underwater vehicle （AUV）, a method based on Singer model of first order time correlation function and strong tracking filter （STF） algorithm is proposed, which combines signal processing method and STF method. This method does not require an accurate mathematical model of the controlled object and has good identification capacity for AUV sensor faults. Simulation and experiment data results for a certain type AUV verify that the proposed method is an efficient and feasible tool for the fault diagnosis of AUV sensor.