Spray deposition characteristics of multi-rotor UAVs equipped with centrifugal nozzles on typical field crops

• Adjuvants can increase droplet deposition but have limited effect on improving uniformity. • The key is to adapt the atomization level to the characteristics of crops. • The centrifugal nozzle has a high atomization effect but may be accompanied by the risk of drift. With the rapid progress in unmanned aerial vehicle (UAV) technology, fields have become the primary application scenarios for spraying. However, research on integrating centrifugal nozzles in UAV spraying applications remains insufficient. This study evaluated and analyzed the spraying performance of a multi-rotor UAV equipped with a centrifugal nozzle for field crops (maize and rice) by examining four factors: adjuvant (Yaobijia), spray volume (15 L·ha −1 , 30 L·ha −1 , 45 L·ha −1 ), atomization level (VF, M, VC), and UAV models (DJI T40, T50). Under the principle of control variables. The droplet deposition results showed that adding adjuvant improved total deposition on the crop canopy (40% for maize and 23% for rice), but the improvement in deposition distribution uniformity was limited. Increasing spray volume could improve canopy droplet deposition but might amplify deposition differences within the canopy. Maize achieved better deposition distribution uniformity at the medium (M) level, and rice demonstrated superior deposition distribution uniformity at the very fine (VF) level for atomization level. Additionally, the study discovered that the variation in deposition between the same series of UAV models (DJI T40 and DJI T50) was not significant ( P >0.05), and deposition of droplet on the leaves underside (UNS) for both UAV models required improvement. Furthermore, droplet size spectrum analysis revealed that over 50% of droplet on both crops were ≤100 μm, confirming the atomization efficiency of centrifugal nozzles. In conclusion, UAV-based spraying with centrifugal nozzles should be optimized by crop type. Rice demonstrates more uniform droplet distribution (lower CV) under ultra-fine atomization mode (VF, 50 μm), while corn adapts better to medium atomization particle size (M, 310 μm). Therefore, operators should select atomization levels according to crop characteristics, while model selection offers greater flexibility.