Research On UAV Resource Scheduling Based On Intelligent Optimization Algorithm

With the development and innovation of science and technology,traditional agriculture has stepped into the digital information age.Wherein,the agricultural precision irrigation is a key link of the modern agricultural production,and exploring the efficient and safe operation mode is now a subject that must be faced.In this thesis,the agricultural scheduling path problem of UAV is studied,conjunctively considering many constraints like the UAV performance,flight endurance,load capacity and the maximum spraying amount of agricultural pesticides.In addition,on the basis of the existing literature and the traditional ant colony algorithm,a fast adaptive dynamic ant colony algorithm(AFD-ACO)is proposed.Aiming at the difficulty of UAV precision irrigation,this thesis proposes a grid processing scheme for the map.Wherein the serial number method is used to jointly monitor the quantitative information of crop pesticide shortage collected by UAV.Considering the constraints of UAV’s maximum load,farthest voyage and energy consumption of various flight modes,the UAV agricultural scheduling method is modelled on a multi-objective programming model,taking UAV’s overall energy consumption and farmland pesticide residues as optimization objectives.Aiming at the defects of slow convergence speed and local stagnation of the traditional ant colony algorithm,the proposed algorithm introduces the elite strategy while choosing the path grid.This way improves and optimizes well the regulation of the pheromone concentration.At the same time,the map is preprocessed,and the pheromone concentration in the initial state is combined with the pesticide shortage in the farmland grid.In addition,the odor diffusion strategy is also employed to greatly improve the convergence speed and search ability of AFD-ACO.Finally,the proposed ant colony algorithm is compared with other algorithms(ACO and EACO)and the algorithm cases are analyzed.in which different map sizes,different crops and different UAV types are applied to different algorithms.The comparison results show that the improved algorithm can obtain the optimal path satisfying the constraint relationships,which is very important for UAV autonomous coverage flight.