Multiobjective routing optimization of mobile charging vehicles for UAV power supply guarantees

•A novel UAV power supply guarantee model is proposed.•UAVs are charged during reconnaissance missions without changing routes.•Two objectives are considered in terms of costs and benefits.•An improved coevolutionary algorithm is proposed to solve the constraint problem.•A Generalized Opposition-Based Learning (GOBL) strategy is employed. A novel power supply guarantee model based on mobile charging vehicles (MCVs) is developed for solving the battery capacity constraint problem of unmanned aerial vehicles (UAVs) in performing long-term and large-scale missions. In this model, the charging vehicles travel to charging points located under the routes of the UAVs to be charged wirelessly, maximizing the assurance of completing reconnaissance missions. This paper focuses on the scheduling and routing of charging vehicles and achieves multisystem cooperation by constructing two submodels, namely, a UAV reconnaissance routing model and an MCV charging routing model. Furthermore, an improved coevolutionary framework for constrained multiobjective optimization problems with a generalized opposition-based learning strategy (OL-CCMO) is proposed to optimize the two objectives of power supply guarantee investment cost and reconnaissance mission time window deviation. Three scenarios of UAV reconnaissance missions with different levels of complexity are built in the case studies, in which the charging vehicle routes are optimized with multiple algorithms, demonstrating the superiority and applicability of the proposed algorithm by comparison.