Quantum Computing Based Collaborative Optimum Mission Allocation Approach for Heterogeneous Multi Unmanned Vehicles

With the developing autonomous driving technologies today, unmanned vehicles - unmanned aerial vehicles (UAV), unmanned ground vehicles (UGV) and unmanned marine vehicles (UMV) - autonomously reach the desired target without human intervention. However, effective and efficient methods for the coordination of these vehicles in multiples need to be developed. Quantum computing has much more potential than classical computers due to its nature. In this study, a quantum-classical hybrid system is proposed to ensure that autonomous transportation unmanned vehicles can assign tasks in a collaborative and fully autonomous manner. Since the constraints of land vehicles navigating on roads and marine vehicles navigating at sea are taken into account in the study, the problem is approached from a different perspective compared to common studies in the literature. First, the QEA algorithm used in the ranking operations was adapted to the problem. Improvements were made to the adapted method to increase its performance. Then, a framework was proposed to operate this method as a quantum-classical hybrid. Significant contributions have been made to the proposed quantum-classical hybrid method by offering advantages in learning the parameters and solution update equations. The problem was solved using classical genetic algorithms (GA), QEA and the proposed method, and their solution performances were compared. It was observed that the proposed method was superior to QEA and produced results very close to those produced by GA. Experimental results confirmed the effectiveness of the proposed method. Implementing the method as a quantum-classical hybrid brings one step closer to the goal of realizing the solution in a quantum environment.