The transferable belief model in ultrasonic map building

Abstract

In this paper we consider the problem of building and updating the map of an a priori unknown environment by using range readings obtained through ultrasonic sensors. A bitmap representation of the world is adopted. In an ideal map, each cell of the bitmap is either empty or occupied by an obstacle. In the case under study, because of the low resolution of the used sensors, evidence on the state (empty/occupied) of a cell is affected by uncertainty that is represented by fuzzy measures. Information provided by measures is combined using Dempster or Smets model to construct an occupancy map of the environment and a comparison between the two approaches is proposed. The map building process is integrated in a sensor-based robot navigation system made up by a repeated sequence of perception and navigation processes.