Scientific paper ID 1886 : 2019/3

MEMS sensors are low price products which determines their widespread use in mobile phones, virtual reality, unmanned vehicle, robots and etc. In the same time theirs output signals have a lot of noise components and are unstable over time. One way to make their characteristics better is to combine data from two or more sensors that are mutually complementary. This is usually done using a complementary filter or Kalman filter. The object of this article are algorithms of complementary filters used to determine the angle of rotation of a child seat in a car. The two algorithms are with different complexity. One of them is with a simpler structure and is suitable for embedding in systems realized by microcontrollers.On the other hand the second one has a more complex structure and describes better the physical processes. Our interest is stopped only on the angle of rotation relative to the longitudinal axis of the vehicle, which can be determined using data from only two sensors - an accelerometer and a gyroscope. These two sensors have complementary frequency characteristics - the accelerometer will determine the change of the angle for a long period of time, while the gyro-data will determine its instantaneous value. The filters are implemented in the Matlab environment and their operation is verified with data obtained from static and dynamic real measurements of the deviation of a child seat, mounted in a car.