Quantum computing based intelligent collaborative control approaches for heterogeneous multi-swarms of unmanned aerial-ground-marine vehicles

Collaborative control approaches for heterogeneous unmanned aerial, ground, marine vehicle (Unmanned Vehicles,UV) swarms are crucial for effective operation in dynamic environments. This thesis proposes a Pareto-based multi-objective route optimization and obstacle avoidance method for unmanned aerial and ground vehicles. The developed method aims to optimize the time to reach the target while minimizing energy consumption and ensuring safe obstacle avoidance in dynamic environments. Additionally, it introduces an innovative optimization framework combining genetic algorithms and costmap-based pathfinding techniques for efficient deployment of UVs in dynamic mission scenarios. This approach, designed to enhance operational efficiency in mission assignment and route planning processes, is optimized to align with criteria specific to different vehicle types. Although unmanned vehicles can operate safely and efficiently without human intervention, complex optimization challenges such as task assignment and route planning emerge in multi-vehicle coordination. Therefore, developing effective and innovative methods for multi-vehicle coordination is essential. In the thesis study, introduces a quantum-classical hybrid system to enable autonomous task assignment for unmanned vehicles in collaboration. The proposed method demonstrates comparable performance to classical genetic algorithms while outperforming quantum-inspired evolutionary algorithms. Additionally, it shows that quantum computing's speed advantage can solve complex problems, which would take longer with classical genetic algorithms, in significantly shorter times. This work offers innovative solutions for task planning, energy efficiency, and obstacle avoidance in dynamic and complex environments for heterogeneous unmanned vehicle swarms. Furthermore, it successfully demonstrates the applicability of quantum computing to task assignment problems, laying the foundation for innovative approaches in this field.