Mission Planning for a Multiple-UAV Patrol System in an Obstructed Airport Environment

This paper investigates using multiple unmanned aerial vehicles (UAVs) to carry out routine patrolling at an airport to enhance its perimeter security. It specifically focuses on mission planning of the system to facilitate efficient patrolling with consideration of local buildings and restricted airspace. The proposed methodology includes three aspects: 1) a vision-based set cover algorithm to construct the patrolling network, 2) an obstructed partitioning-based clustering algorithm for recharging station placement, and 3) a mixture integer quadratic programming (MIQP) algorithm to plan routes for UAVs minimizing the maximum idle time through out all patrolling waypoints. The main contribution of this work is that it provides a comprehensive mission planning solution for UAVs persistently patrolling in a complex environment characterized by blocked vision and restricted airspace. The proposed methodology is evaluated through intensive simulations in the context of the Cranfield Airport scenario.