PeerAppear: A distributed geospatial index supporting collaborative world model construction and maintenance

Abstract This paper addresses the problem of scalable location-aware distributed indexing to enable the leveraging of collaborative effort for the construction and maintenance of world-scale maps and models. These maps and models support numerous activities including navigation, visual localization, persistent surveillance, and hazard or disaster detection. We approach a solution through the creation of PeerAppear, a location-aware framework for peer-to-peer indexing, search and retrieval. Due to the dynamic nature of the world, the problem of constructing and maintaining relevant world-scale models generally requires significant effort to be spent on mapping. PeerAppear offers a decentralized solution which enables the leveraging of collaborative effort through the implementation of a peer-to-peer middleware framework which automates the indexing and sharing of sensed geospatial information captured and stored in the local repositories of participants. The PeerAppear network achieves scale through a Kademlia-like overlay network which indexes data based on location by adapting Google’s S2 hierarchical geographic segmentation scheme to a globally addressable distributed geographic table. Our communications primitives allow search queries to be formed and executed, enabling the discovery of information published in a specified geographic area. An evaluation of the framework is presented demonstrating excellent retrievability of published data, logarithmic efficiency and global scalability.