Helicopter–UAVs search and rescue task allocation considering UAVs operating environment and performance. (May 2022)

• Record Type: Journal Article
• Authors: Zhang, Ming; Li, Wei; Wang, Mengmeng; Li, Songrui; Li, Boquan
• Is Part Of: Computers & industrial engineering. Volume 167(2022)
• Publication Date: 2022-05
• Subjects: Search and rescue -- UAV -- Task allocation -- Improved binary bat algorithm -- Task balance; Engineering -- Data processing -- Periodicals; Industrial engineering -- Periodicals; 620.00285
• DOI: 10.1016/j.cie.2022.107994
• Languages: English
• ISSNs: 0360-8352

Abstract

Highlights: UAV energy consumption was analyzed considering low-altitude wind and terrain. UAV release position selection model was constructed and solved via IBBA. The principal component analysis was used to determine the search and rescue level. Search and rescue task allocation model was established and solved using NSGA-II. The examples show that the proposed model can effectively solve the problem. Abstract: The task allocation of unmanned aerial vehicles (UAVs) is important in UAV search and rescue to guarantee an effective and orderly search and rescue activity. However, few studies have investigated the influences of the UAV operating environment and performance on search and rescue task allocation. Considering the influences of low-altitude wind and terrain on UAV energy consumption and performance, UAV release position (from helicopter) selection and task allocation models are proposed in this study. Given that the detection zone of UAVs is affected by topographic factors, the principal component analysis method was used to determine the search and rescue level at each search and rescue point and the UAV hovering endurance, which was determined using clustering analysis. Considering the influence factors of UAV battery energy consumption, such as UAV performance and low-altitude wind, an UAV release position selection model was constructed and solved via the improved binary bat algorithm which exhibited enhanced calculation accuracy comparing with three other algorithms. According to the release position planning result, a multi-objective (minimization of the total search and rescue cost and number of UAVs and multi-UAV task balance) optimization model was established and solved using the non-dominated sorting genetic algorithm-II. Finally, the experimental results indicated that the solution through multiple objectives had no cost advantage comparing with those under the single objective, but exhibited an evident advantage in task completion time through sensitivity analysis of UAVs task allocation problem.