In a recent study published in the journal Electronics, researchers proposed leveraging smartphones (SPs) as mobile assistance devices for unmanned aerial vehicles (UAVs) in task execution based on analysis incentives. They formulated a multi-objective optimization problem aimed at balancing UAV costs with SP execution utility and developed a multi-objective mutation-immune bat (MOMIB) algorithm to address this challenge.
Study: Smartphone-Assisted UAVs Enhance Disaster Relief Efficiency. Image Credit: Panchenko Vladimir/Shutterstock.com
They also introduced Quality of Service (QoS) coefficients to address the performance requirements of various task types. The goal was to develop a cost-effective and efficient system that enhances UAV capabilities and ensures reliable task execution.
Background
UAVs are versatile and adaptable devices equipped with various modules, including information acquisition, communication, and computing modules. They are commonly deployed in disaster areas to collect data and transmit it to disaster relief centers (DRCs) or cloud servers. However, their lightweight design limits battery capacity, which constrains flight time and endurance. Additionally, UAVs' computing modules may face challenges processing large volumes of real-time data. To address this, UAVs can offload some of their computing tasks to devices with greater processing power.
SPs are widely used portable devices that people carry daily, offering strong computing capabilities, diverse communication options, and widespread availability. In disaster situations, many SPs with idle computing resources are available. These SPs can support UAVs by handling computation-intensive tasks, allowing UAVs to conserve battery power for data collection while providing incentives for the SPs' processing contributions.
About the Research
In this paper, the authors designed and developed a system model with three objects: UAVs, SPs, and DRCs. UAVs collect tasks within their coverage areas and offload some tasks to nearby SPs. SPs are located in DRCs, which help residents based on the disaster's severity, such as shelter and relief supplies.
The study considered two disaster levels: mild and severe. In mild disasters, SPs can move in and out of DRCs, while in severe disasters, SPs remain relatively fixed. The researchers modeled disaster scenarios and introduced QoS coefficients to assess the performance needs of different tasks. Each task was described by data size, maximum latency, and QoS requirements.
SPs received incentives for task execution based on the base incentive, data size, and total delay. The authors established a multi-objective optimization problem to minimize UAV overhead while maximizing SP utility, including constraints for SP power limits and task delay tolerance.
To solve this, they developed the MOMIB algorithm, which combines the bat algorithm (BA) with the clone immune algorithm. The BA mimics bat echolocation to generate and update the population of solutions, while the clone immune algorithm simulates biological immune responses to perform high-frequency mutation and selection. Additionally, the MOMIB algorithm uses a population adjustment evaluation algorithm to rank and trim solutions based on optimization indicators. The MOMIB algorithm aims to find a balanced solution known as Pareto optimality.
Research Findings
The new algorithm's performance was evaluated through simulations and compared with three baseline approaches: the stochastic task scheduling (STS) algorithm, the emergency task priority scheduling (ETP) algorithm, and the original bat algorithm (OB). The experiments covered various conditions, including different numbers of tasks, SPs, iterations, and data environments.
The outcomes showed that the developed algorithm outperformed the baseline approaches in reducing UAV overhead and increasing SP utility. It demonstrated effective optimization and convergence across different scenarios, proving its effectiveness and stability. Additionally, the algorithm achieved fast convergence and a good balance between multiple objectives.
Applications
The proposed technique has valuable implications for disaster relief operations, where UAVs and SPs can work together for data collection and processing. Using SPs as mobile assistance devices, UAVs can conserve battery power and extend flight time, improving disaster relief efficiency and effectiveness. Incentivizing SPs encourages participation, increases available computing resources, and reduces task completion time. Additionally, introducing QoS coefficients ensures that different tasks meet their performance requirements, enhancing service quality and user satisfaction.
Conclusion
In summary, the novel SP-based task scheduling approach for UAV networks proved effective, reliable, and efficient for disaster relief. It outperformed baseline methods in optimization, stability, and robustness across various environments, enhancing UAV network efficiency and service quality, offering valuable support for disaster relief efforts.
Moving forward, the authors suggested future research directions, including exploring task collaboration in multi-UAV settings, addressing communication delays and interference between UAVs and SPs, and incorporating more realistic factors into the system model.
Journal Reference
Li, L.; Wang, Z.; Zhu, J.; Ma, S. Smartphone-Based Task Scheduling in UAV Networks for Disaster Relief. Electronics 2024, 13, 2903. DOI: 10.3390/electronics13152903, https://www.mdpi.com/2079-9292/13/15/2903
Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.