基于卷积神经网络的MEMS传感器随机误差补偿方法

Random error compensation method of MEMS sensor based on convolutional neural network

为解决MEMS传感器随机误差补偿方法中漂移误差补偿能力较低,导致补偿模型角速率误差补偿效果较差等问题,提出基于卷积神经网络的MEMS传感器随机误差补偿方法。采用多方位、多层次的数据采集方法,将MEMS传感器放置在多个不同实验点车辆上,在行驶过程中采集多个信息数据。对原始数据进行环境误差消除处理,以保证误差数据的准确度,分析MEMS传感器随机误差类型,进行误差整体标定,实现误差补偿。实验结果表明,所提方法漂移误差补偿能力得到有效提高,补偿模型角速率误差补偿效果较好。

In order to solve the problem that the ability of drift error compensation in MEMS sensor random error compensation method is low,resulting in poor compensation effect of angular rate error of compensation model,a random error compensation method of MEMS sensor based on convolution neural network is proposed.The multi-directional and multi-level data acquisition method is adopted.The MEMS sensors are placed on vehicles at different experimental points to collect multiple information data during driving.In order to ensure the accuracy of the error data,the random error types of MEMS sensors are analyzed,and the overall error calibration is carried out to realize the error compensation.The experimental results show that the drift error compensation ability of the proposed method is effectively improved,and the angular rate error compensation effect of the compensation model is better.