Integrating Safety-Aware Route Optimisation and Run-Time Safety Monitoring in Controlling Swarms of Drones

Swarm of drones are increasingly deployed to perform a variety of critical missions such as surveillance, rescue in disaster areas etc. To guarantee success of a mission, the controlling software should pursue two goals. Firstly, it should ensure safety, i.e., guarantee collision avoidance. Secondly, it should prevent a premature depletion of the batteries of the drones by minimizing their travel paths. In this paper, we propose an approach that combines run-time safety monitoring and high performance evolutionary algorithm to predict dynamically emerging hazards. High performance of the route calculation algorithm allows us to ensure that the routes of drones are dynamically adjusted to avoid collisions while maintaining efficiency. The benchmarking of the proposed approach validates its efficiency and safety.