Autonomous Agricultural Sprayer using Machine Vision and Nozzle Control

This paper proposes a modular system of precision agriculture to automate sprayers, optimizing the application of pesticides through a robotic system based on computer vision and individual nozzle on/off control. The system uses low-cost equipment such as Arduino boards, solenoid valves, pressure and flow sensors, smartphone, webcam, and Raspberry Pi. The motivation is to reduce the amount of pesticides applied in crops, not just for potential savings for the farmers, but also for environment protection issues, as well as for food safety. The system can be used in any crop planted in rows such as onion, soybean, corn, beans, and rice. The results show that our system can detect lines in plantations and can be used to retrofit conventional boom sprayers, so it is an important step to develop a kit capable of upgrade a conventional sprayer to a fully autonomous robotic sprayer even at affordable cost in the context of small and medium size farms.