Abstract
Though space travel is technologically developed, bad weather makes path planning inefficient. When an unmanned aerial vehicle looks for an optimal path, it is hard to carry out a mission from unidentified obstacles. In the existing method, the path planner does not work with multiple UAVs. Also path optimization is considered as a big challenge for multiple UAV movement. The proposed work, uses Multiple Swarm Fruit fly optimization with the Q-Learning method for path planning in multiple UAVs. The proposed MSFOA divides the total fruit fly swarms into sub-swarms with multi-job also improves the searching area. In addition, the Q Learning-based path selection strategy is used to optimize global and local searches during the evolutionary process. Whereas the offspring competitive method is used to enhance the level of use of each computation result and to facilitate the transmission of information across different fruit fly sub-swarms. The proposed method MSFO-QL for path planning reduces costs and finds the path more efficiently. Computational results show that the suggested MSFO-QL can handle the limited UAV path planning with minimum worst selection with 280 on case 1 and 232 on case 2. Proposed is more efficient and robust than the existing optimization techniques.