基于Kalman滤波和神经网络的MEMS陀螺温度漂移补偿

MEMS陀螺温度漂移严重影响系统的测量精度。传统的BP神经网络建模补偿容易使权值和阈值陷入局部极小值,导致网络训练失败。陀螺输出信号中的高频噪声也会影响模型精度。针对上述问题,该文提出一种Kalman滤波结合粒子群算法(PSO)优化BP神经网络的MEMS陀螺温度漂移补偿方法。首先对陀螺进行了温度漂移测试实验,然后采用Kalman滤波对实验数据进行降噪,最后建立陀螺温度漂移模型,从而实现温度漂移的补偿。实验结果表明,采用该方法补偿后MEMS陀螺在不同温度下的输出方差降低了65.09%,与传统的BP神经网络相比补偿精度明显提高。

Temperature Drift Compensation of MEMS Gyroscope Based on Kalman Filter and Neural Network

The system measurement accuracy of the MEMS gyroscope is seriously affected by temperature drift. The traditional BP neural network modeling compensation can easily make the weights and thresholds fall into the local minimum,which leads to the failure of network training. High frequency noise in the gyro output also affects the model accuracy. Aiming at the above problems,a method of compensating temperature drift of MEMS gyroscope based on Kalman filter and particle swarm optimization(PSO) is proposed to optimize BP neural network. Firstly,the temperature drift test of the gyroscope is carried out,then the experimental data are de-noised by Kalman filter. Finally,the temperature drift model of the gyroscope is established to compensate the temperature drift. The experimental results show that the output variance of the MEMS gyroscope is reduced by 65.09% at different temperatures,and the compensation accuracy is improved significantly compared with the traditional BP neural network.