Reasoning with multiple-source information in a possibilistic logic framework

This paper addresses the problem of merging uncertain information in the framework of possibilistic logic. It presents several syntactic combination rules to merge possibilistic knowledge bases, provided by different sources, into a new possibilistic knowledge base. These combination rules are first described at the meta-level outside the language of possibilistic logic. Next, an extension of possibilistic logic, where the combination rules are inside the language, is proposed. A proof system in a sequent form is given, sound and complete with respect to the possibilistic logic semantics. Possibilistic fusion modes are illustrated on an example inspired from an application of a localization of a robot in an uncertain environment.