The research, published in Smart Agricultural Technology, tackles a stubborn blind spot in precision agriculture. Sensor-guided sprayers have become remarkably good at adjusting how much liquid they release: ultrasonic sensors, LiDAR, and stereo vision systems can map a tree canopy in real time and pulse individual nozzles on and off through solenoid valves driven by pulse-width modulation signals. Variable-rate sprayers built around stereo vision have cut pesticide use by 72.6 to 80.5 percent and reduced ground losses by 57.6 percent compared with conventional constant-rate machines. But while the liquid side of spraying has grown intelligent, the air side has barely changed in decades.

Conventional orchard sprayers rely on axial fans that blast a constant, high-volume stream of air to carry droplets into dense foliage. That fixed airflow is a compromise at best. Directed at a sparse canopy, it blows droplets straight past the leaves and into the atmosphere; aimed at dense foliage, it lacks the momentum to push droplets into the interior where pests hide. Mechanical baffles and louvers have been tried to modulate the output, but they respond slowly, offer coarse control, consume large amounts of energy, cannot adapt to shifting winds, and can even damage crops with excessive airflow. Because canopy density changes over a season and wind conditions shift minute to minute, no fixed setting can consistently balance deposition against drift.

The new work builds on an electric air-assisted spraying system, or EAAS, developed earlier by Hongyoung Jeon and Heping Zhu. The system uses a tower design with vertically stacked electric fans that can swing air velocity from zero to 11.3 meters per second in roughly three seconds, delivering air horizontally and parallel to the ground rather than in the turbulent radial pattern of conventional fans. Earlier field tests showed how sensitive spray performance is to fan speed: low air assistance of 3.4 meters per second increased average canopy deposition on full-foliage apple trees by 24.6 percent over unassisted spraying, while maximum output of 11.3 meters per second cut deposition by 23.2 percent and inflated drift potential by up to 401.4 percent. The message was clear: more air is not better air.

To turn that insight into a working controller, the team led by Matthew J. Herkins ran a systematic campaign of wind tunnel experiments. Their test rig paired a 397-millimeter electric axial fan with a flat-fan spray nozzle, both governed by an Arduino microcontroller and a solid-state relay that adjusted fan speed in 5 percent duty-cycle increments, yielding outlet air speeds from 1.86 to 10.96 meters per second. Inside a low-speed wind tunnel capable of laminar airflow up to about 10 meters per second, they sprayed an artificial Ficus tree placed 2.05 meters downwind. Three foliage conditions, leafless, half-foliage, and full-foliage, corresponded to leaf area index values of 0.00, 1.39, and 2.77, spanning the range measured on apple trees across a growing season. Wind speeds from 0.7 to 4.3 meters per second and wind directions sweeping from directly favorable to directly adverse completed the test matrix.

Deposition was measured with a fluorescent tracer, Brilliant Sulfaflavine, collected on stainless steel meshes at three canopy heights, while water-sensitive papers recorded spray coverage and downwind samplers captured airborne drift. The results revealed a striking pattern: when wind blew with the spray direction at 270 degrees, deposition peaked at the lowest fan speed tested and fell as air speed rose, a consequence of leaf layering, in which high-velocity airstreams flatten foliage against itself and shield interior surfaces. As wind direction swung oblique, the optimum fan speed climbed steadily, reaching the maximum 10.96 meters per second at 305 degrees. Denser canopies demanded more air to overcome aerodynamic drag, but only when crosswinds deflected the plume. Statistical analysis confirmed the hierarchy: fan air speed alone explained 41.2 percent of deposition variance, and its interaction with wind direction accounted for another 38.3 percent, dwarfing every other factor.

From this dataset the researchers built empirical regression models that predict the optimal fan speed from wind speed, wind direction, and canopy density. For favorable wind directions, they developed five models corresponding to drift tolerance thresholds of 0, 2.5, 5, 10, and 15 percent, each defining how much deposition a grower will accept in exchange for less drift. The models used physics-informed features, including perpendicular and parallel wind efficiency terms that decompose the wind vector relative to the spray trajectory, and a saturating leaf area transformation. Cross-validation was rigorous: five-fold and ten-fold schemes, plus leave-one-angle-out and leave-one-wind-out tests. Under ten-fold validation, coefficients of determination ranged from 0.783 to 0.865, and the final refitted 0 percent tolerance model achieved an R-squared of 0.894 with a root mean square error of just 1.10 meters per second.

Adverse winds demanded a different strategy. When spraying into wind at 90 or 118 degrees, the optimum was near-maximum fan output in 14 of 15 measurable conditions, and at 4.3 meters per second under the oblique adverse direction, no deposition was recorded at any fan speed. The team fitted a single third-order polynomial that returns near-maximum air assistance throughout this envelope, keeping the controller logic continuous rather than predictive. The authors caution that the models are empirical and valid only within the experimental calibration domain; conditions outside it require recalibration, not extrapolation, and denser canopies beyond a leaf area index of 2.77 still need field validation.

The implications reach well beyond the wind tunnel. Because the models are simple enough to run on a sprayer’s onboard controller, they could be folded directly into the EAAS tower, where each vertically stacked fan module would apply the equations using wind and canopy inputs local to its own height. The researchers outline the road ahead: implementing the models on a field sprayer, adding travel speed to capture machine-generated turbulence, exploring machine-learning approaches to smooth abrupt fan transitions, and linking the system to local weather stations so the sprayer anticipates changing conditions rather than lagging behind them. Reliable real-time wind estimation from a moving platform remains a prerequisite for deployment.

If those steps succeed, the payoff could be substantial for specialty crops such as tree fruits, nuts, and vines, which are among the most intensive pesticide consumers per acre in U.S. agriculture. A sprayer that dials its fan down when the wind cooperates and the canopy is thin, and pushes it to maximum only when crosswinds demand it, would cut the chemical and energy waste baked into every fixed-airflow pass. In an era when agriculture faces mounting pressure to protect workers, rural communities, and waterways from pesticide exposure, teaching a sprayer to feel the wind may prove one of the most consequential upgrades precision farming has delivered in years.

Subject of Research: Automatic air assistance control for electric air-assisted agricultural spraying systems

Article Title: Development of an automatic air assistance control model for an electric air-assisted spraying system

Article References: Herkins, M. J., Jeon, H., Zhu, H., Zhao, L., You, K., Marzano, S.-Y., & Ozkan, E. (2026). Development of an automatic air assistance control model for an electric air-assisted spraying system. Smart Agricultural Technology, 15, Article 102619. https://doi.org/10.1016/j.atech.2026.102619

Image Credits: AI Generated

DOI: Not provided

Keywords: precision agriculture, pesticide drift, air-assisted sprayer, variable-rate application, leaf area index, wind tunnel, canopy deposition, spray coverage, regression model, orchard spraying, electric fan control, smart agricultural technology

The stakes are considerable. Weeds cost soybean growers more than 30 percent of their yield in severe cases, translating to an estimated 33 billion U.S. dollars in annual global losses. Worse, infestations often appear in scattered patches rather than uniform blankets, meaning a broadcast herbicide application treats vast stretches of weed-free ground. Depending on how weeds are distributed, spot spraying can cut herbicide use by as much as 66 percent when weeds cluster in small patches, though savings drop to roughly 10 to 20 percent when weeds spread evenly across a field. Existing ground robots can perform targeted spraying, but they struggle in wet soil, on slopes, or in small and irregularly shaped fields, and they must traverse large weed-free areas to reach isolated outbreaks.

The Arkansas team, led by Md Nurul Azmir and Cengiz Koparan, took a deliberately different route from the deep-learning mainstream. Instead of training a convolutional neural network to distinguish soybean from weed pixel by pixel, an approach that demands large annotated datasets and costly retraining for every new crop, growth stage, or field, their system quantifies total vegetation coverage in each image using a training-free segmentation method based on the Hue, Saturation, Value (HSV) color space. Pixels falling within calibrated HSV ranges are classified as vegetation, and coverage is simply the proportion of vegetation pixels in the image footprint. The clever twist lies in the reference: vegetation coverage measured from weed-free soybean plots at the same growth stage serves as a baseline, and any coverage exceeding that baseline is interpreted as a proxy for weed pressure.

The hardware is equally pragmatic. A MAPIR Survey3N multispectral camera, recording near-infrared, red, and green bands, feeds images to an NVIDIA Jetson Orin Nano, a compact edge-computing board with a 1024-core GPU and 8 GB of memory that draws at most 25 watts. When estimated weed pressure crosses a predefined threshold, the Jetson issues a pulse-width modulation (PWM) signal to an electronic speed controller, which drives a 12-volt brushless pump rated at 3.0 liters per minute and feeding two TeeJet nozzles. The entire sensing-to-actuation chain, from image capture through segmentation, decision logic, and pump activation, runs onboard, independent of the drone’s flight controls, with the DJI AGRAS MG-1S serving purely as the aerial carrier.

Field validation took place at the Milo J. Shult Agricultural Research and Extension Center near Fayetteville, where twenty soybean plots were planted in July 2024 with no herbicides applied, allowing natural weed populations to develop. Five plots were kept weed-free by hand to establish the crop reference. Across 1,000 images collected 14 and 24 days after planting, weed-free soybean coverage averaged 3.6 percent at the early date and 14.2 percent at the later one, but with substantial plot-to-plot and image-to-image variability, coefficients of variation reaching above 50 percent. That variability matters: a spray threshold set at the simple mean of weed-free coverage would trigger spraying on healthy crops, so the researchers recommend basing thresholds on the average of plot-level maximum coverage values, with a safety margin.

Threshold sensitivity analysis revealed a growth-stage-dependent tradeoff familiar to site-specific weed management. At 14 days after planting, when weed-free and weedy coverage distributions overlapped heavily, raising the threshold from 5 to 7.5 percent cut potential false spray activation from about 21 percent to essentially zero, but also collapsed the spray-trigger rate in weed-infested plots from 52 percent to 15 percent. By 24 days after planting, the distributions had separated enough that a 25 percent threshold produced only 4.8 percent false activation while still triggering sprays on 88.3 percent of weed-infested footprints, with a potential treatment-footprint reduction of 32.6 percent. In other words, the system’s discriminative power improves as the canopy develops, a finding that could let growers time their spot-spraying missions for maximum selectivity.

Indoor tests under controlled lighting confirmed the control logic end to end. Using green reference markers to simulate crops and weeds, the system correctly kept its pump off over bare background and crop-only configurations, then activated automatically when simulated weeds pushed total vegetation coverage to 14 percent and the weed-pressure proxy to 64.9 percent. Across 147 spray-actuation trials, the system matched the visually observed pump state 143 times, an accuracy of 97.3 percent, with precision of 99.1 percent and only three false negatives. System-estimated weed pressure tracked marker-based reference values with a coefficient of determination of 0.91, though a positive bias of 7.3 percentage points, traced to edge pixels around marker boundaries, suggests room for refinement in threshold calibration.

Field performance was more modest but still encouraging. Comparing system-derived vegetation coverage against manual ImageJ segmentation of the same images yielded a coefficient of determination of 0.69, a mean absolute error of 2.20 percentage points, and a slight negative bias of 2.02 percentage points. The gap between indoor and field agreement reflects the messy reality of real canopies: mixed crop-weed arrangements, variable illumination, shadows, and soil background all conspire against fixed color thresholds. During supplementary UAV flight tests, the sprayer sometimes failed to activate, likely because shadows and changing solar angle pushed computed coverage below threshold, or because motion blur from flight and rotor downwash degraded image quality. The authors are candid that these environmental factors remain the principal obstacle to reliable airborne operation.

The team is equally clear about what the study does not yet show. Spray deposition, droplet coverage, off-target drift, actual herbicide savings, and end-to-end sensing-to-actuation latency were not quantified, and pump-flow calibration alone does not characterize nozzle-level spray performance, which depends on operating pressure, droplet size, and effective spray width. Future validation plans call for water-sensitive papers placed at detected spray and no-spray locations to verify placement accuracy, alongside georeferenced detection outputs. Payload and battery constraints on small UAVs will also shape how much ground each mission can cover and how much spray volume it can carry.

Even with those caveats, the study makes a compelling case that vegetation coverage relative to a crop-only baseline is not merely a mapping metric but an actionable control variable, closing a loop that much weed-sensing research leaves open. By sidestepping species-level classification, the framework runs on modest hardware without labeled data, making it adaptable to new crops, growth stages, and geographies simply by recalibrating HSV thresholds and re-measuring a weed-free reference. The authors envision the approach extending to multi-platform systems in which drones scout rapidly and hand off targeted intervention to ground robots with greater payload and endurance. For an industry wrestling with herbicide resistance, environmental contamination, and the economics of treating weed-free ground, a drone that sprays only what needs spraying is a proposition whose time, and technology, may finally be converging.

Subject of Research: Adaptive UAV-based spot spraying for weed management in soybean using vegetation coverage thresholds

Article Title: Development of an adaptive sprayer control system for UAV-based spot spraying in soybean

Article References: Azmir, M. N., Tagoe, A., Runkle, B. R., Wang, D., Burgos, N. R., & Koparan, C. (2026). Development of an adaptive sprayer control system for UAV-based spot spraying in soybean. Smart Agricultural Technology, 15, Article 102589. https://doi.org/10.1016/j.atech.2026.102589

Image Credits: AI Generated

DOI: 10.1016/j.atech.2026.102589

Keywords: precision agriculture, UAV spraying, spot spraying, weed management, soybean, HSV segmentation, edge computing, variable-rate application, herbicide reduction, vegetation coverage, machine vision, sustainable farming