基于LSTM神经网络的机载光纤陀螺温度冲击误差补偿技术

Temperature Shock Error Compensation Technology for Airborne Fiber Optic Gyroscopes Based on LSTM Neural Networks

环境温度冲击会降低机载光纤陀螺的性能,从而影响飞行器导航和姿态控制精度。在光纤陀螺误差机理研究基础上,本文提出一种基于长短期记忆(LSTM)神经网络的光纤陀螺温度误差补偿模型。该模型通过LSTM网络对光纤陀螺的零偏和标度因数进行实时预测和校正,提高光纤陀螺的测量精度。试验结果表明,在温度冲击下,LSTM预测模型补偿后的标度因数误差小于30ppm,零偏稳定性比常规的线性拟合补偿模型提高0.0034(°)/h。这意味着输出更准确地反映实际角速度值,陀螺仪的零偏漂移更小,输出更接近于零值。动态试验中转台输入为20(°)/s时,LSTM补偿后陀螺输出稳定在19.999~20.001(°)/s区间内,相较于陀螺原始输出误差降低0.008(°)/s。通过LSTM预测模型补偿,能够在环境变化、外部扰动或传感器故障时,通过陀螺仪提供更可靠的数据支持,维持飞行器的稳定性和安全性。

The measurement accuracy of the onboard fiber optic gyroscopes could be reduced by environmental temperature shocks,consequently impacting the flight accuracy of the aircraft.A temperature error compensation model based on long short-term memory(LSTM)neural networks was proposed in this paper to improve the measurement accuracy of fiber optic gyroscopes under temperature shock.The zero bias and scale factor of the fiber optic gyroscope were predicted and corrected in real-time using the LSTM network,improving its measurement accuracy.Experimental results showed that under temperature shock,the scale factor error was compensated by the LSTM prediction model,which was less than 30ppm.The zero bias stability was improved by 0.0034(°)/h compared with the conventional linear fitting compensation model.In dynamic experiments,when the input of the turntable was set to 20°/s,the gyroscope output was stabilized in the range of 19.999~20.001(°)/s after LSTM compensation,and the error of the gyroscope original output was reduced by 0.008(°)/s.The changes of the zero bias and scale factor of the airborne fiber optic gyroscope under temperature shock were more effectively compensated by the LSTM network.The stability of the inertial navigation of aircraft was enhanced.