With the increasing application of plant protection unmanned aerial vehicle (UAV) in precision agriculture, the numerical simulation methods for the development of the downwash flow field of the plant protection UAV and the deposition and drift process of droplets affected by the downwash flow field have achieved rapid and diversified development, but the advantages, disadvantages, scope of application, and verification of each method still lack a systematic review. This article discusses the inviscid model, computational fluid dynamics model and lattice Boltzmann model (LBM) respectively. The advantage of the inviscid wake vortex model based on the vortex element method is that the calculation process is simple. Moreover, integrated with the most widely used aerial spray drift prediction software AGricultural DISPersal (AGDISP), it can be a promising way to do real-time UAV spray drift prediction. But due to lack of viscosity and turbulence models, the droplet deposition and drift simulation accuracy of inviscid model is relatively lower than other models. The computational fluid dynamics (CFD) model includes the finite volume method (FVM) and the finite difference method (FDM). The FVM in the computational fluid dynamics model has high robustness and can be applied to the simulation of various complex environments. Many commercial CFD software are based on FVM and achieved a fast development in aerial spray modeling recently. However, the FVM is greatly affected by the quality of the mesh, and its commonly used upwind style has limited accuracy (second-order accuracy). Under the same mesh density, it is easier to generate artificial dissipation when simulating the rotor tip vortex than the finite difference method. As a result, the simulated rotor tip vortex dissipation speed is much faster than the actual situation. Compared with the FVM, the structured grid used in the FDM is easier to construct a high-order precision numerical format. Which can reach 4-5 orders of accuracy, and with adaptive grid technology, FDM can simulate the evolution of rotor tip vortex with high temporal and spatial accuracy, and can reproduce the typical flow structure development process of the real rotor downwash flow field. However, it also has problems such as high grid structure requirements and excessive computing power requirements. LBM has advantages in computing three-dimensional flow field problems with complex boundary conditions and non-stationary moving objects. However, there are still shortcomings in its functional diversity and completeness. The accuracy of the numerical models mentioned above still needs field test and indoor experiment such as high-speed Particle Image Velocimetry (PIV)/ Phase Doppler Interferometry (PDI) method to verify and optimize. The authors finally pointed out the future direction of plant protection UAV application simulation and verification. [ABSTRACT FROM AUTHOR]