Performance evaluation of novel distributed coverage techniques for swarms of flying robots

This paper focuses on the coverage of specific Zones of Interest that can change dynamically over time by using a swarm of flying robots. The mobility of the flying devices is achieved by the design of two distributed and local algorithms. The first algorithm is based on Particle Swarm Optimization (PSO) and Virtual Forces Algorithm (VFA). We modify the classical PSO approach to propose a totally distributed algorithm, which only requires the flying robots to receive local information from the neighbors to update their velocity and trajectory (PSO-S). This new distributed version of the PSO is combined with a distributed version of the Virtual Forces Algorithm. The second algorithm is a distributed implementation of the VFA (VFA-D). To the best of our knowledge, these two approaches are novel in their distributed character, scalability and implementability on resource-constrained devices. We show that the proposed algorithms are reactive, i.e. able to capture in an effective fashion the events happening within the field even if the position of the events changes over time. To show the effectiveness of the proposed techniques, we perform extensive simulations to compare both the PSO-S and the VFA-D schemes with a centralized version of the VFA. Simulations show the good performance in terms of coverage and traveled distance as well as the high reactivity of both PSO-S and VFA-D when the ZoI changes.