Coevolutionary Optimization for Multi-UAV Path-Planning

The paper proposes a novel parallel optimization method to solve cooperative path-planning problem for a network of multiple unmanned aerial vehicles (UAVs) for disaster management. The path-planning problem addressed here is framed in the settings of an optimization problem involving eight objectives. The problem is solved by independent optimization of the objectives in parallel. The scope of parallel optimization enhances the efficiency to thoroughly explore individual multi-modal fitness landscapes. It in turn increases the robustness of the movements of multiple UAVs in complex terrain. Firefly Algorithm (FA) is selected as the base algorithm because of its established performance. Moreover, the decisive control parameters of FA are adapted based on their performance history with an aim to balance between the explorative and exploitative capabilities. Experimental results demonstrate the superiority of the proposed algorithm over its competitors.