Abstract: Abstract: The liquid quantity data is one of the most important information to spraying UAV (unmanned aerial vehicle). During the aerial spraying, the liquid quantity in the pesticide tank is dynamic, the operators have always to paid attention to the liquid quantity so that they can make suitable decisions for the flight controlling. Therefore, the liquid quantity monitoring method for pesticide tank is the most important in UAV spraying operation. However, other special characteristics need to be evaluated in order to achieve real-time monitoring the amount of pesticide liquid of spraying UAV, including unstable liquid surface, variation of physical and chemical properties of different pesticides, shapes of different tanks, the anti-corrosion requirements, tank sizes, etc. In this paper, we aimed to explore the feasible method for liquid quantity monitoring of pesticide tank in spraying UAV. Based on the above characteristics, we developed a solution called dual-pressure liquid quantity monitoring system, which consisted of double pressure sensors for liquid-level monitor, wave filter for liquid level turbulence (i.e.va combination of the median filter and the moving average filter), liquid level correction for aircraft on a tilt (based on a gyroscope), and conversion model for liquid level and liquid quantity. In order to verify the feasibility of the proposed method, a prototype of liquid monitoring device was developed, and the relevant verification test was conducted to test its performance. The experimental results showed that there was a significant (p < 0.05) linear negative correlation between the liquid level and pressure difference (between ambient atmospheric pressure and sealed air chamber pressure caused by liquid level), and the correlation coefficient R2 was 0.9989. It showed that the dual-pressure method can effectively eliminate the influence of ambient atmospheric pressure changes on the measurement accuracy and stability. By using the proposed wave filter, the C.V. (coefficient of variation) of liquid level data decreased from 28.45% to 12.27%, indicating that the proposed wave filter can filter off the liquid level turbulence effectively. By using liquid-level correction for aircraft on a tilt, the error was reduced to 0.05 from -1.09 hPa when the aircraft tilt up to 30°. Therefore, the proposed tilt correction method can effectively eliminate the error caused by aircraft body tilt. In order to test the performance of prototype of liquid quantity monitor in the actual spraying operation, the test based on a spraying UAV, 3WDM4-10, was conducted. The test included three different flying attitudes, forward, backward and turn around at the edge of field. When the payload of the pesticide tank was 2, 4 and 6 L, the mean liquid quantity data out by the liquid quantity monitor were 1.985, 3.942, and 5.984 L, the relative errors of the liquid quantity monitor were 0.75 %、1.45%、0.77%, the RMS error was 0.182, 0.199 and 0.180 L, respectively. The results demonstrated that the output of the proposed liquid quantity monitor is stable and reliable in different flying attitude.