Mission Arrangement Optimization with Improved PSO for Multiple Plant Protecion UAVs in Heterogeneous Farmlands

Abstract

Multiple plant protection Unmanned Aerial Vehicles (UAVs) are often used as a team in protection operation in a large area farmland plant nowadays. Mission arrangement is the first problem should be solved before the operation. In this work, we focus on promoting the operation efficiency of the multiple UAVs, systematically proposed Optimal Spraying Mission Assignment Problem (OSMAP), and try to construct the united problem framework in the mission arrangement of multiple plant protection UAVs. Firstly, the mathematical model of multiple UAVs was established and the OSMAP is defined. Secondly, the acquisition of the Normalized Difference Vegetation Index (NDVI) data of farmlands is proposed, and the decision method of pesticide volume based on a farmland’s NDVI map was considered, which heterogeneous features of the sub-areas of a farmland. Thirdly, an improved Particle Swarm Optimization (PSO) is proposed, and numerical simulations for algorithm analysis are carried out in MATLAB. It is determined that the proposed PSO is more efficient and accurate than the classic PSO. Finally, a mission arrangement test is implemented with a 4-P20-1-P30-UAV system with farmland NDVI data was carried out to validate the effectiveness of the proposed algorithm in spraying operation. Test results proved that the proposed PSO performed well, which took only 63.98% of the operation time taken by the general mission arrangement method.