SISTEMA DE LOCALIZACIÓN AUTÓNOMA PARA ROBOTS MÓVILES BASADO EN FUSIÓN DE SENSORES PROPIOCEPTIVOS.

The main objective of this study is to develop an autonomous localization system capable of delivering better position estimates compared to an exclusively odometer system by means of a sensor fusion algorithm. A mobile robot travels a pre-programmed path to provide sensory data to the system. A fusion architecture is define that works with odometers, accelerometers and gyroscope data. The robot movement model, the measurement model and the sensory data are using an Extended Kalman Filter. The results show that in all the cases that were evaluated the system records an improvement of 38% compared to a standard deterministic localization system. The data show that the θ variable is the most influential in the process. In conclusion, the results satisfy the stated objective, nevertheless, it can be improved by incorporating additional sensors and adjusting the uncertainty matrices R and Q. [ABSTRACT FROM AUTHOR]