Design and Test of Target Spraying Robot Based on Visual Recognition
In order to solve the problem that the traditional large-scale indiscriminate spraying of pesticides will cause pollution to the environment and easily produce pesticide residues and endanger health,this paper designed a target spraying robot based on visual recognition to solve the problem of pesticide waste caused by conventional pesticide appli-cation.The Raspberry Pi,CMOS camera and servo are used to collect and process images,and they are equipped on the rubber wheeled mobile platform to realize the design of the target spraying robot.The target spraying robot adopts the lat-eral application method,using the VP110-015 nozzle with the 12V water pump to achieve precise spraying,and the verification distance of 1.1m to collect 320×240 pictures(.jpg format)meets the processing requirements,and the de-lay relay resistance value of 100 Ω can meet the requirements of precise spraying,the single spraying volume of the wee-ding robot is 0.002 L,and the single spraying covers the three-dimensional space of 0.004 m3.When the current ad-vance speed is0.10 m/s,the recognition rate reaches more than90%,and the application rate of liquid medicine can be reduced by 73.34%by using visual recognition and target spraying.The results of this study are of great significance for reducing the use of pesticides,improving the target rate,and realizing the sustainable development of agriculture.
target spray robotvisual recognitiontarget sprayRaspberry Pi
万方数据
维普
