DESIGN, DEVELOPMENT, AND EVALUATION OF A NOVEL HYDRAULIC SPRAY SYSTEM FOR MACHINE VISION-BASED REAL-TIME TARGET APPLICATION OF PESTICIDES ON BOOM SPRAYERS

Abstract

A novel spraying mechanism assembly designed for machine vision–based, real time target application of pesticides on self propelled boom sprayers was developed and evaluated. The research addressed key limitations in modern precision spraying – such as nozzle valve actuation latency, boom pressure instability, and spray accuracy at higher ground speeds. The study investigates how 3D spray nozzles and a controller area network (CAN)–based feed forward pressure stabilization system can enhance timing accuracy and spraying at prescribed application rate. A comprehensive experimental investigation was conducted to characterize and optimize spray atomization parameters, including droplet size, velocity, breakup behaviour, spray coverage, and collision outcomes, across system pressures ranging from 275 to 482 kPa. Results show that 3D nozzles generate backward inclined spray patterns that effectively compensate for mechanical nozzle valve latency during real-time operation. A system pressure of 275 kPa emerged as the optimal pressure due to coarser droplets, minimized drift risk, favourable droplet coalescence, and reduced pumping power requirements. A CAN compatible feed forward pressure stabilization technique was developed to regulate boom pressure in anticipation of nozzle activation events, outperforming conventional feedback based systems by reducing pressure deviations by up to 63% and shortening pressure settling times by up to 5.7 fold. The system was prototyped on a 60 nozzle target spraying boom as an add-on system and validated through laboratory and field experiments, including tests on artificial and real Colorado potato beetle (CPB) targets. Field results demonstrated accurate real time actuation, consistent droplet placement, and practical integration of machine vision detection with hydraulic and electronic control subsystems. Overall, the research establishes an effective, scalable framework for deploying real time, target specific pesticide application technology that reduces chemical use, improves timing accuracy, and advances the feasibility of precision pest management in commercial agriculture