Robot Map Building From Sonar Sensors and DSmT

摘要

Knowledge acquisition in map building presents characteristics of uncertainty and imprecision. Especially in the course of building grid map using sonar, this uncertainty is especially severe. Jean Dezert and Florentin Smarandache have recently proposed a new information fusion arithmetic (DSmT) whose greatest merit is to deal with uncertainty and conflict of information. In this paper, based on the arithmetic of DSmT, we can fuse information of different reliable degree for homogeneous or heterogeneous sensors. Then we established the belief model for sonar grid map, and constructed the generalized basic belief assignment function (gbbaf). Pioneer II mobile robot served as the experiment platform, and 3D Map was built based on DSmT online. At last, this paper established a firm foundation on studying dynamic unknown environment and multi-robots building map together and SLAM (simultaneous localization and mapping).