Privacy Risk Mitigation Strategies for Drone Package Delivery

Abstract

Uncrewed aerial vehicles (UAVs), or drones, are increasingly used to deliver goods. In an emerging business model, a drone operator partners with multiple businesses to offer drone delivery as a service. Due to regulations requiring drones to broadcast position information, this business model results in a privacy risk: Third-party observers may use broadcast drone trajectories to link customers to the vendors from which they order, with a wide range of potential consequences. We propose a probabilistic definition of privacy risk based on the likelihood of inferring which customer receives a delivery from which vendor. Next, we quantify these risks and evaluate the impacts of the number of orders, drone capacity, decoy vendors, and delivery lime time requirements on privacy. We then discuss how privacy risk may be integrated into the vehicle routing problem or explicitly optimized on its own. Finally, we show the geographical dependence of the trade-off between privacy and efficiency.