Design Of Visual Servo Control System For Spraying Robot Of Hyphantria Cunea Larva Nets

The Hyphantria cunea is a worldwide quarantine pest.Since its introduction into the Dandong area of Liaoning Province in 1979,its hazard area has been increasing year by year,causing serious damage to crops,trees,fruit trees,etc,which has seriously threatened,long-term sustainable development of forestry resources.In order to effectively deal with the serious damage caused by the Hyphantria cunea to the ecological environment,the visual servo control system design of the Hyphantria cunea Larva Nets spray robot is designed in this paper.The design can adapt can to the irregularity of the Hyphantria cunea Larva Nets distribution and control PWM to achieve accurate variable spraying according to the depth information of the Nets.The system can effectively improve the utilization of pesticides,reduce the pollution of pesticides to the environment,and at the same time reduce labor intensity.The main research contents are as follows:(1)Visual design of visual servo control system: Image of Hyphantria cunea Larva Nets was acquired by using camera,and the image was denoised and enhanced by median filtering and gradient sharpening.The candidate region was obtained by selective search method,and the region was scored by using the Convolutional Neural Network(CNN)model according to the set threshold,and the nets was extracted from the picture.(2)Control design of visual servo control system: According to the irregular distribution of the Hyphantria cunea Larva Nets,a progressive scan method is designed to locate the precise position of the screen in the image and a coordinate system was established to calibrate the actual position of nets.the PWM was controlled to achieve precise variable spraying according to the intensity of the nets.(3)System experiment: By using the experimental prototype and the simulation tree to carry out the accurate variable spraying experiment,the qualitative relationship between duty cycle and flow rate and pressure was determined;secondly,the light intensity and wind speed experiments were carried out to conclude the influence of light intensity and wind to the system image acquisition and precise variable spraying.Theresults show that the visual servo control system achieves precise variable spraying,and the coverage of the target reaches over 90%.