Nature-Inspired Drone Swarming for Wildfires Suppression Considering Distributed Fire Spots and Energy Consumption

• Main Authors: Alsammak I.L.H., Mahmoud M.A., Gunasekaran S.S., Ahmed A.N., Alkilabi M.
• Other Authors: 57220190775
• Format: Article
• Published: Institute of Electrical and Electronics Engineers Inc. 2024
• Subjects: Random walk algorithm; stigmergy; swarm intelligence; UAVs; wildfires suppression; Agricultural robots; Antennas; Drones; Energy utilization; Fires; High resolution transmission electron microscopy; Intelligent robots; Job analysis; Losses; Numerical methods; Particle swarm optimization (PSO); Random processes; Swarm intelligence; Aerial vehicle; Classification algorithm; Particle swarm; Particle swarm optimization; Random walk algorithms; Stigmergy; Swarm optimization; Task analysis; Unmanned aerial vehicle; Wildfire suppression; Forestry

Wildfires are among the biggest problems faced worldwide. They are increasing in severity and frequency, causing economic losses, human death, and significant environmental damage. Environmental factors, such as wind and large forest areas, contribute to the fire spreading over multiple fire spots, all of which grow continuously, making fire suppression extremely difficult. Therefore, fire spots should be coverage simultaneously to contain the spread and prevent coalescence. Therefore, this study presents a new model based on the principles of nature-inspired metaheuristics that uses Swarm Intelligence (SI) to test the effectiveness of using an autonomous and decentralized behaviour for a swarm of Unmanned Aerial Vehicles (UAVs) or drones to detect all distributed fire spots and extinguishing them cooperatively. To achieve this goal, we used the improved random walk algorithm to explore the distributed fire spots and a self-coordination mechanism based on the stigmergy as an indirect communication between the swarm drones, taking into account the collision avoidance factor, the amount of extinguishing fluid, and the flight range of the drones. Numerical analysis and extensive simulations were performed to investigate the behaviour of the proposed methods and analyze their performance in terms of the area-coverage rate and total energy required by the drone swarm to complete the task. Our quantitative tests show that the improved model has the best coverage (95.3%, 84.3% and 65.8%, respectively) compared to two other methods Levy Flight (LF) algorithm and Particle Swarm Optimization (PSO), which use the same initial parameter values. The simulation results show that the proposed model performs better than its competitors and saves energy, especially in more complicated situations. � 2013 IEEE.