We study the average Age of Information (AoI) in a system where physical sources produce independent discrete-time updates that are each observed by several sensors. We devise a model that is simple, but still capable to capture the main tradeoffs. Two sensor scheduling policies are proposed to minimize the AoI of the sources; one in which the system parameters are assumed known, and one in which they are learned. Both policies are able to exploit the common sensor information to reduce the AoI, resulting in large reductions in AoI compared to common schedules.