Practical Considerations of IMU Data Generator

Abstract

Research in inertial navigation field constantly has to deal with sensor data and sensor errors. Therefore, an IMU (Inertial Measurement Unit) data generator is mandatory to save the time of logging data and isolate the sensor errors. The IMU data generator is also more flexible in generating inertial data of different types of trajectory. An IMU generator contains a gyroscope data generator and an accelerometer data generator. In general, the gyroscope and acceleration data are generated by taking derivative of attitude and position. There are three different ways to generate gyroscope data depending on how the attitude is presented. This paper shows the practical problems, which alway present in the implementation of the IMU data generator. That is noise appearing in the derivative of input data. The method to remove derivative noise shows that the noise is rejected with maximum shape preservation of trajectory. Further, an investigation is carried out to show which methods should be used to generate the gyroscope data for maximum input motion preservation.