Numerical calibration for 3-axis accelerometers and magnetometers

Magnetometers and accelerometers are sensors that are now integrated in objects of everyday life like automotive applications, mobile phones and so on. Some applications need information of acceleration and attitude with a high accuracy. For example, MEMS magnetometers and accelerometers can be integrated in embedded like mobile phones and GPS receivers. The parameters of such sensors must be precisely estimated to avoid drift and biased values. Thus, calibration is an important step to correctly use these sensors and get the expected measurements. This paper presents the theoretical and experimental steps of a method to compute gains, bias and non orthogonality factors of magnetometer and accelerometer sensors. This method of calibration can be used for automatic calibration in embedded systems. The calibration procedure involves arbitrary rotations of the sensors platform and a visual 2D projection of measurements.