Leveraging Collective Run-Time Adaptation for UAV-Based Systems

UAV-based systems are systems that are composed of a team of drones, various devices (like movable cameras, sensors), and human agents, which collaborate each other to accomplish defined missions. Since humans are constituent part of these systems, UAV-based systems are both mission-critical and safety-critical. Moreover, these systems are requested to operate in potentially unpredictable and unknown environments. A model of the environment describing, e.g. obstacles, no-fly zones, wind and weather conditions might be available, however, the assumption that such model is both correct and complete is often wrong. In this paper, we describe a novel approach for managing the run-time adaptation of UAV-based systems. Our approach is based on a generic collective adaptation engine that addresses collective adaptation problems in a decentralized fashion, operates at run-time, and enables the addition of new entities at any time. Moreover, our approach dynamically understands which parts of the system should be selected to solve an adaptation issue. The feasibility and scalability of the approach have been empirically evaluated in the context of a private company surveillance scenario.