In this paper, we investigate the use of a mobile, autonomous agent to update knowledge bases containing statements that lose validity with time. This constitutes a key issue in terms of knowledge acquisition and representation, because dynamic data need to be constantly re-evaluated to allow reasoning. We focus on the way to represent the time- validity of statements in a knowledge base, and on the use of a mobile agent to update time-invalid statements while planning for “information freshness” as the main objective. We propose to use Semantic Web standards, namely the RDF model and the SPARQL query language, to represent time-validity of information and decide how long this will be considered valid. Using such a representation, a plan is created for the agent to update the knowledge, focusing mostly on guaranteeing the time-validity of the information collected. To show the feasibility of our approach and discuss its limitations, we test its implementation on scenarios in the working environment of our research lab, where an autonomous robot is used to sense temperature, humidity, wifi signal and number of people on demand, updating the knowledge base with time- valid information.
展开▼
