(57) The invention relates to a method (100) for calibrating at least one on micro-electromechanical
systems based inertial sensor (50), comprising the following steps:
- Providing (101) input data, the input data comprising at least sensor data, the
sensor data being specific for the at least one inertial sensor (50),
- Predicting (102) a working point by a machine-learning model based on the provided
input data, the working point being specific for a parametrization of the at least
one inertial sensor (50),
- Carrying out (103) iterative adjustment steps for a fine adjustment of the working
point, the adjustment steps comprising a comparison of an effect of a currently set
value with a reference value for each inertial sensor (50), the currently set value
being set based on the predicted working point, the number of iterations of the adjustment
steps being flexible and depended on the comparison,
- Providing (104) a final parameter field for the at least one inertial sensor (50)
based on the predicted working point or the adjustment steps for the fine adjustment.
|

|