ASABE Technical Library - Abstract

Abstract.

Precision target spraying based on internet remote control has been studied in this paper. The machine vision method was used to control disease during plants growth stage in greenhouse. A 3G controller was equipped on a hanging spraying machine to realize the remote control of the sprayer, and machine vision sensors were equipped on it to realize target spraying. To limited massive instructions in remote control of the spraying machine, BP neural network was used to predict the time-delay of network. PID control method was introduced to put forward a control algorithm with time-delay estimate and compensation, which could improve the dynamic performance and the stability during the remote control of spraying machine. Meanwhile, AES encryption algorithm was used to improve the security performance in the remote control. An on-board machine vision system was constructed to achieve the precision disease control of plectranthus tomentosa in greenhouses. A disease feature extraction algorithm in HSV color space was established. Then the disease infected region could be accurately discriminated. This information was provided to the spraying control system as the basis for the precision spraying. The results showed that its average time-delay was less than 500ms. The correct recognition rate of disease was 93.3%, which based on the identification of 300 potted plectranthus tomentosas. The target error of the spraying machine was less than 5cm. The results showed that this method could meet the requirements of the remote precision target spraying in the greenhouse.