Occupancy grid mapping for mobile robot using sensor fusion

Abstract

Sensor data fusion using more than one senor such as sonar sensors fusion reduces uncertainties generated from a single sensor. To learn the environment using more than one sensor information, an accurate sensor model as well as a reasonable sensor fusion methodology is needed. In this work, the Moravec-Elfes sonar model for occupancy grid representation and the recursive Bayes update rule in sensor fusion is applied. The environment of the mobile robot may be highly uneven for that only one type of sensors are not enough. Hence to increase the sensor accuracy to a great extent, the information obtained from two sonar and two laser sensors are combined for identifying different shape of objects.