A Comparison of the Effect of Sonar Grid Map Building Based on DSmT and DST

Information acquired in map building presents characteristics of uncertainty, imprecision and even high conflict, especially in the course of building grid map using sonar. Here a new method of information fusion - DSmT (Dezert-Smarandache theory) recently developed from DST and Bayesian theory was introduced to apply to managing and dealing with this uncertainty. Aiming to static environment, classic DSm mathematic modal was established by constructing the general basic belief assignment function, so that the global evidence grid map is built by applying the classic DSm rule. Pioneer II virtual mobile robot served as an experiment platform. By comparing the numbers of acquiring-information spot, the quality of grid map building and the time in need of finishing the global map building, better fusion results than that with DST are gotten. Especially when dealing with high conflictive information from sonar grid map building, DSmT was more effective and successful than DST. In short, this study not only found a new method for building map under static environment, but also supplied with a theory foundation for us to further apply hybrid DSmT to dynamic unknown environment and multi-robots' building map together