Abstract:This paper investigates key technologies of a quantum-classical hybrid algorithm-based simulation platform for unmanned aerial vehicle (UAV) swarm mission planning, to address the issues of large-scale problem, complicated constraints, and local optimality in conventional algorithms. A modular architecture is developed for swarm cooperative mission simulation, consisting of mission management, classical path planning, quantum-classical hybrid optimization, conflict detection and post-processing, and simulation execution. A scalable, modular modeling approach is proposed to decouple individual UAV models and swarm task logic. A multi-constraint joint mission-path planning model is established, where task allocation and path optimization are collaboratively solved by combining NSGA?II and a distributed auction mechanism. Task assignment and path selection are formulated as a quadratic unconstrained binary optimization problem, and a quantum-classical hybrid framework is constructed using the quantum approximate optimization algorithm and a classical optimizer.Software-in-the-loop simulations in typical 3D scenarios verify that the platform supports scalable swarm simulation with a frame rate above 50 Hz. The joint planning scheme improves task completion rate by approximately 8% and reduces path length by about 9% compared with phased planning. The quantum-classical hybrid algorithm achieves superior objective values in combinatorial optimization. The results validate that the platform provides effective support for algorithm research and engineering verification of UAV swarm mission planning.
