Semantic Modeling and Reconstruction of Drones' Trajectories

Research on semantic trajectories' modeling, analytics, and visualization has been conducted for a wide range of application domains. In contrast to raw trajectories, semantically annotated trajectories provide meaningful and contextual information to movement data. Unmanned Aerial Vehicles (UAVs), also known as drones, are becoming more and more widely used in modern battlefields as well as in search and rescue (SAR) operations. Semantic trajectories can effectively model the movement of swarms of drones towards enabling decision makers/commanders to acquire meaningful and rich contextual information about Points of Interest (PoI) and Regions of Interest (RoI) that will eventually support simulations and predictions of high-level critical events in the real field of operations. The goal of this paper is to present our position related to the semantic trajectories of swarms of drones, towards proposing methods for extending MovingPandas, a widely used open-source trajectory analytics and visualization tool. Such an extension is focused on the semantic modeling of drone trajectories that are automatically reconstructed from geo-tagged data, such as photographs taken during a flight mission of a swarm of UAVs, where its flight plan or real-time movement data have been either lost or corrupted, or there is a need for semantic trajectory cross-validation.