Sensor Fusion for a mobile robot

Abstract

Consulta en la Biblioteca ETSI Industriales (9112)

[ES] As outlined in Section 1.1, every sensor has an internal weakness. On one hand, the gyroscope tends to drift and on the other hand, the accelerometer is a ected by vibrations making it unstable, however, the accelerometer does not drift. The objective of this thesis is to combine an accelerometer and a gyroscope using two di erent methods to obtain the estimated angle of a mobile inverted pendulum. The rst method implemented is a Complementary Filter where the information from every sensor has a xed weight in the estimated angle data. This lter uses a high pass lter and a low pass lter to prevent the sensor from the drift, the noise and the vibrations. The second method used is a Kalman Filter. This method is much more di cult to implement but gives the best results as it is based on the minimization of the errors from the true signal value.