Swarm Intelligence Optimization of UAV Routing with Simultaneously Stochastic Pick-up and Delivery during COVID-19

During the COVID-19 pandemic, it is imperative to distribute daily supplies to residents in lockdown communities and collect their household garbage. Unmanned Aerial Vehicle (UAV) logistics offers a contactless solution, but the limitations of load and maintenance costs must be considered. Therefore, it is crucial to plan UAV transportation routes effectively to minimize overall transportation costs. Additionally, the stochasticity of household garbage quality further complicates UAV delivery. We address this issue by formulating the problem as vehicle routing problem with simultaneous stochastic pickups and deliveries (VRPSSPD) and develop a range of swarm intelligent optimization algorithms (SIOAs) to solve it. Our experiments demonstrate that the Artificial Bee Colony (ABC) algorithm and Shuffled Frog Leaping Algorithm (SFLA) are well-suited to this practical problem. Moreover, to address the long convergence time of SFLA, we propose the General Center SFLA based on Roulette Wheel Selection (RWS-GC-SFLA) to improve it. Our experimental results show that RWS-GC-SFLA can improve the convergence time while ensuring solution quality. Finally, we apply RWS-GCSFLA to a practical instance case of UAV logistics in Guangzhou communities and achieved satisfactory results.