Smart sprayer for weed control using an object detection algorithm (YOLOv5)

An experimental variable-rate sprayer designed for liner applications was tested by comparing its spray deposit and coverage, and droplet density inside canopies of six nursery liner varieties with constant-rate applications. Spray samplers, including water sensitive papers (WSP) and nylon screens, were mounted inside tree canopies to collect spray deposit and coverage from variable- and constant-rate (555 and 1,110 L/ha) applications. Models for estimating spray volume saving of the sprayer compared to a constant-rate and Tree-Row-Volume (TRV) based rate applications were developed for various liner canopy sizes and tree spacing. The accuracy of the model was validated with the field test data. For the liner trees tested, the variable-rate sprayer delivered 151 to 359 L/ha application rates while the conventional constant-rate application required 1,110 L/ha. Due to substantially lower spray output, the variable-rate application had relatively lower but sufficient spray deposit, coverage, and droplet density than the constant-rate applications. Also, the variable-rate application had significantly less variations in spray deposit within the canopies of different size trees. Therefore, the newly developed variable-rate sprayer would bring great reductions in pesticide use and safeguard the environment for the nursery liner production.

An experimental variable-rate sprayer designed for liner applications was tested by comparing its spray deposit, coverage, and droplet density inside canopies of six nursery liner varieties with constant-rate applications. Spray samplers, including water-sensitive papers (WSP) and nylon screens, were mounted inside tree canopies to collect spray deposit and coverage from variable-rate and constant-rate (555 and 1,110 L ha-1) applications. Models for estimating spray volume savings of the variable-rate sprayer compared to constant-rate and tree-row-volume (TRV) rate applications were developed for various liner canopy sizes and tree spacings. The accuracy of the model was validated with the field test data. For the liner trees tested, the variable-rate sprayer delivered 151 to 359 L ha-1 application rates, while the conventional constant-rate application required 1,110 L ha-1. Due to substantially lower spray output, the variable-rate application had lower (but sufficient) spray deposit, coverage, and droplet density than the constant-rate applications. The variable-rate application also had significantly less variations in spray deposit within canopies of different-size trees. Test results showed that the variable-rate sprayer reduced spray volume by up to 86.4% and 70.8% compared to the 1,110 L ha-1 and TRV-based rate applications, respectively, while the model-estimated savings were up to 94.6% for the 1,110 L ha-1 application and 57.7% for the TRV-based rate applications. Therefore, the newly developed variable-rate sprayer would bring great reductions in pesticide use and safeguard the environment for nursery liner production.

Multiple-row container-grown trees require specially designed sprayers to achieve efficient spray delivery. A five-port air-assisted sprayer with both automatic and manual control modes was developed to discharge adequate spray deposition inside multiple-row trees. The sprayer resulted from integration of a high-speed laser-scanning sensor with a sophisticated nozzle flow control system, an embedded computer with a touch screen, a Doppler speed sensor, a specially-designed algorithm and an air-assisted sprayer base. It was able to detect target tree presence and measure target tree size, shape and leaf density. The sprayer then controlled the spray output of each nozzle to match tree structures. The sprayer was tested for its sprayer deposition quality inside canopies in a four-row sterling silver linden (Tilia tomentosa 'Sterling Silver' Moench) field and another six-row northern red oak (Quercus rubra L.) field. Tests were conducted with the sprayer in variable-rate application (VRA) and constant-rate application (CRA) modes. The average spray deposit on foliage of trees was 1.37±0.47 μL cm−2 from VRA and 1.29±0.42 μL cm−2 from CRA in linden, and was 2.15±0.57 μL cm−2 from VRA and 2.72±0.94 μL cm−2 from CRA in red oak, respectively. In comparison, spray coverage on foliage of trees was 19.8±3.0% from VRA and 20.9±4.3% from CRA in the linden trial, and was 27.9±3.7% from VRA and 30.5±5.4% from CRA, respectively, in the red oak trial. The newly developed air-assisted sprayer in both VRA and CRA modes would be able to discharge adequate spray deposition inside multiple-row tree plants while conserving pesticide. Index words: environmental protection, pest control, pesticide, precision sprayer, spray coverage. Species used in this study: sterling silver linden (Tilia tomentosa 'Sterling Silver' Moench), northern red oak (Quercus rubra L.).

Highlights A real time stereo vision controlled variable rate sprayer for specialty crops was developed. The stereo vision system of the sprayer detected outdoor trees with similar canopy profiles under travel speeds ranging from 3.2 to 8 km h-1. Canopy volume measurements of the sprayer were impacted by lateral distances between the sprayer and the tree center and travel speeds. The sprayer required less than 200 ms from tree canopy detection to spray decisions. The sprayer achieved spray volume reductions from 72.6% to 80.5% compared to constant rate spray application. Abstract. A real time variable rate sprayer controlled by a stereo vision system was developed to increase the accuracy of spray applications and reduce the use of crop protection products. The sprayer was designed to detect tree canopies and calculate its volume using depth images from the stereo vision system, and discharge corresponding spray volumes every 200 ms through the embedded software in the graphical user interface. The sprayer was evaluated in an apple orchard at different travel speeds (3.2 to 8.0 km h-1) for its performance in detecting canopy and measuring its volume. In addition, spray volume, deposition, and coverage of the variable rate application of the sprayer were evaluated against a constant rate application. Test results showed that the sprayer detected visually similar tree canopies during the evaluations, although its canopy volume measurements deviated from manually measured canopy volume from 0.11 to 0.83 m3 due to lateral position changes of the sprayer. The sprayer adjusted duty cycles of pulse width modulated valves to accurately spray the intended volume for detected canopies (0.073 to 0.083 L m-3) and only used spray volumes of 19.5% to 26.7% compared to a constant rate spray application (338 L ha-1). The constant rate spray application generally had more spray deposition and coverage in tree canopies than the variable rate sprayer, as expected since its spray volume was approximately 3.7 times higher. However, the mean spray depositions from the constant rate spray application were significantly varied (p=0.05) by tree sizes, while the variable rate spray application achieved statistically equivalent mean spray depositions regardless of tree sizes. The stereo vision controlled sprayer offers a cost-effective real-time variable rate spray option for growers with the potential to perform other tasks by using image processing algorithms while applying crop protection products. Keywords: Automation, Canopy volume, Crop protection, Depth image, Orchard, Precision agriculture, Real-time application.

<abstract> <bold>Abstract.</bold> A newly developed LiDAR-guided air-assisted variable-rate sprayer for nursery and orchard applications was tested at various travel speeds to compare its spray deposition and coverage uniformity with constant-rate applications. Spray samplers, including nylon screens and water-sensitive papers (WSP), were mounted inside tree canopies to measure spray deposits and coverage. Experiments were conducted with six different size tree species in commercial nurseries and travel speeds ranged from 3.2 to 8.0 km h-1. Test results showed that variable-rate sprayer was able to adjust spray outputs to match canopy structures and travel speeds. For the variable-rate application, there were no significant variations in spray deposits and coverage for each tree species among the travel speeds of 3.2, 5.6 and 8.0 km h-1. The variable-rate application had significantly lower variations in spray deposits within canopies of different size trees than the constant spray application.

An experimental variable rate nursery sprayer was developed to adjust application rates for canopy volume in real time. The sprayer consisted of two vertical booms integrated with ultrasonic sensors, and variable rate nozzles coupled with pulse width modulation (PMW) based solenoid valves. A custom–designed microcontroller instructed the sensors to detect canopy size and occurrence and then controlled nozzles to achieve variable application rates. A spray delivery system, which consisted of diaphragm pump, pressure regulator and 4-cycle gasoline engine, offered the spray discharge function. Spray delay time, time adjustment in spray trigger for the leading distance of the sensor, was measured with a high–speed camera, and it was from 50 to 140 ms earlier than the desired time (398 ms) at 3.2 km/h under indoor conditions. Consequently, the sprayer triggered 4.5 to 12.5 cm prior to detected targets. Duty cycles of the sprayer were from 20 to 34 ms for senor–to –canopy (STC) distance from 0.30 to 0.76 m. Outdoor test confirmed that the nozzles were triggered from 290 to 380 ms after detecting tree canopy at 3.2 km/h. The spray rate of the new sprayer was 58.4 to 85.2% of the constant application rate (935 L/ha). Spray coverage was collected at four areas of evergreen canopy by water sensitive papers (WSP), and ranged from 1.9 to 41.1% and 1.8 to 34.7% for variable and constant rate applications, respectively. One WSP area had significant (P < 0.05) difference in mean spray coverage between two application conditions.

The present study was undertaken to determine whether a biventricular bypass system operated in an independent variable rate (VR) mode can maintain the entire circulation. Two pusher-plate pumps which incorporated the Hall effect position sensors were used to bypass the right and left ventricles in 10 sheep under fibrillation. The flow distributions of the pump output to the carotid and renal arteries were investigated every 6 h using ultrasonic blood flow meters for 24 h in 5 animals, and the controllability of the VR mode was evaluated in 5 long-term experiments. The carotid artery flow ratio to the pump output decreased significantly from 4.7 +/- 0.8% before the bypass to 2.7 +/- 0.9% after 24 h. However, the renal artery flow ratio did not change throughout the experiments. In the long-term experiments, the animals were kept alive from 3 to 48 days (mean 15.6 days). The mean pump output had been maintained at more than 90 ml/min/kg for the first 7 days. After the surgery, the pump driving conditions were not readjusted in any experiment. The results indicate that the biventricular bypass system operated in the independent VR mode automatically maintains the entire circulation at a satisfactory level.

Highlights Stereo vision controlled variable rate sprayer reduced the spray volume by 44% to 99.6% for the tree canopy volumes from 2.2 to 0.03 m3. The variable rate sprayer increased the ratio of a spray deposit on trees to a spray volume by 16.1% compared to a constant rate sprayer. The application of the variable rate sprayer had 57.6% less ground loss compared to a constant rate application. The travel speeds of the sprayer had no significant impacts in spray deposition, coverage, or ground losses of the applications. Abstract. A prototype of a novel stereo vision controlled real-time variable rate sprayer was evaluated in an apple orchard in order to mitigate the risks derived from the use of crop protection products applied at a conventional constant rate in specialty crops. The effects of its travel speed (from 3.2 to 8.0 km h-1) on spray volume reduction, canopy deposition and coverage, and ground loss were assessed. Test results demonstrated that the travel speed did not influence spray deposition, coverage, or ground losses. The variable rate sprayer reduced the spray volume by more than 44% for the canopy volumes detected in the orchard (&amp;lt;2.2 m3) in comparison to the constant rate spray application. Overall spray volume reductions were 79.1%, 73.8%, 71.0%, and 69.8% when the sprayer traveled at 3.2, 4.8, 6.4, and 8.0 km h-1, respectively, compared to the constant rate applications at a travel speed of 8.0 km h-1. In addition, the variable rate application of the sprayer increased the crop interception, a ratio of a spray deposition to a spray volume, by 16.1% compared to the constant rate spray application when traveling at 8 km h-1. Moreover, the sprayer reduced total ground loss by 57.6% compared to the constant rate application, which would minimize adverse impacts of pesticide applications to non-target organisms and unnecessary pesticide losses to the environment. The results of this study suggest that the prototype sprayer has significant potential as a cost-effective solution for sustainable specialty crop spray applications. Keywords: Crop protection, Depth image, Machine vision, Orchards, Precision agriculture, Pulse width modulation.

Recent trends in herbicide applications show a preference among peanut growers for using lower carrier volumes and coarser-droplet nozzles. Field-scale studies using commercial application equipment were conducted in 2021 and 2022 to investigate the influence of carrier volume and nozzle type on spray coverage, droplet density, and weed control in peanut. The study treatments consisted of target carrier volumes of 94, 117, and 140 L ha-1, with each volume applied using three different nozzles – XRC, AIXR and TTI – to attain different droplet sizes. Spray coverage and droplet density data were collected during herbicide applications. Weed control was recorded after herbicide applications and peanut yield was measured at harvest. Spray coverage improved with an increase in carrier volume from 94 to 140 L ha-1. The AIXR nozzles provided comparable (2021) or improved (2022) coverage than the XRC nozzle while reduced coverage was observed for the TTI nozzle during both years. Droplet density was greatest for the XRC nozzle followed by the AIXR and TTI nozzles but was not impacted by carrier volume. Despite noticeable differences in spray coverage and/or droplet density, weed control and peanut yield were not affected by carrier volume and nozzle type. Overall, these findings suggest that peanut growers may observe reduced spray coverage for herbicide applications at low carrier volumes and/or when using nozzles that produce large (Ultra Coarse) droplets, but this effect may not directly translate into reduced herbicide efficacy or peanut yield in most fields with low to moderate weed pressure. Future studies should investigate the influence of spray parameters (carrier volume, nozzle type, etc.) on weed management in fields with varying weed pressures and different weed sizes.

Three non-ionic adjuvants, Agral, Silwet, and Greemax, at three concentrations, were applied on apple leaves with the use of hollow cone nozzles (TR) and air-induction nozzles (ID) to verify the assumption that adjuvants may improve spray coverage obtained by coarse droplets, and thereby ensure both satisfactory application quality and an environmental advantage. Spray coverage and droplet density were measured on both sides of the leaves. The adjuvants enhanced the spray coverage when applied at a certain concentration level. In general, the adjuvant coverage produced by the ID nozzles equaled the pure water coverage produced by the TR nozzles, thereby showing the adjuvants’ potential to compensate for the lower spray coverage usually obtained by coarse spray. A higher spray coverage was obtained on the lower side of leaves, which is discussed in terms of leaf surface properties. In the experiment with the mixture of Silwet and the fungicide Delan (dithianon), the product interacted with the adjuvant, resulting in the reversed picture of spray coverage and droplet density on the upper and lower leaf sides compared to the results obtained for the adjuvant alone. The combination of coarse spray nozzles with adjuvants may reduce environmental pollution without compromising the quality of spray applications in fruit growing.

To optimize pesticide applications to the canopies of deciduous perennial crops, spray volume should be adjusted throughout the year to match the changes in canopy volume and density. Machine-vision, computer-controlled, variable-rate sprayers are now commercially available and claim to provide adequate coverage with decreased spray volumes compared with constant-rate sprayers. However, there is little research comparing variable- and constant-rate spray applications as crop characteristics change throughout a growing season. This study evaluated spray volume, spray quality (e.g., coverage and deposit density), and off-target spray losses of variable- and constant-rate sprayers across multiple phenophases in an apple ( Malus domestica ) orchard and a grape ( Vitis vinifera ) vineyard. The variable-rate sprayer mode applied 67% to 74% less volume in the orchard and 61% to 80% less volume in the vineyard. Spray coverage (percent), measured by water-sensitive cards (WSC), was consistently greater in the constant-rate mode compared with the variable-rate mode, but in many cases, excessive coverage (i.e., over-spray) was recorded. The variable-rate sprayer reduced off-target losses, measured by WSC coverage, up to 40% in the orchard and up to 33% in the vineyard. Spray application deposit densities (droplets per square centimeter) on target canopies were typically greater in variable-rate mode. However, the deposit densities were confounded in over-spray conditions because droplets coalesced on the WSC resulting in artificially low values (i.e., few, very large droplets). Spray efficiencies were most improved early in the growing season, when canopy density was lowest, demonstrating the importance of tailoring spray volume to plant canopy characteristics.

Comparison of within canopy deposition for a solid set canopy delivery system (SSCDS) and an axial–fan airblast sprayer in a vineyard

Highlights The effectiveness of four nozzles with flat-fan spray pattern types was evaluated to assess their ability to deliver spray droplets inside the soybean canopy in a wind tunnel equipped with a moving boom. The effect of wind speed on the spray coverage, deposition, and airborne drift of droplets discharged from the four evaluated nozzles was investigated under the wind speeds of 0, 2.4, and 5.1 m s-1. Spray deposition at the bottom part of canopies was significantly lower than at the upper part. Nozzles that produced ultra-coarse droplets reduced airborne spray drift risks. The dense soybean canopies and overlapping foliage of crops limit the spray droplet penetration capability of the hydraulic nozzles. ABSTRACT. Adequate spray deposits and coverage in the middle and lower parts of the canopy are essential to protect soybean crops from disease and insect attacks. Additionally, uncontrollable weather conditions can influence the effectiveness of spray applications. Thus, the objective of this research was to evaluate the effect of the wind speeds of 0, 2.4, and 5.1 m s -1 on the spray coverage and deposition discharged from travelling nozzles with different droplet sizes under wind tunnel-controlled conditions. An open-circuit wind tunnel, equipped with a moving spray boom placed 0.5 m above the top of the soybean canopy and operating at a speed of 0.9 m s -1 , was used in this study. To achieve the application rate of 150 L ha -1 , pulse width modulation solenoid valves were coupled with the nozzles at a duty cycle of 19% and an operating pressure of 276 kPa. Four types of same color-coded 110° flat-fan spray nozzles (XR11004, TTJ6011004, AITTJ6011004, and AIXR11004) were used to determine the effects of spray droplet size spectra Keywords: Centrifugal fan, Laminar flow, PWM nozzle, Spray deposition, Spray drift, Traveling boom, Turbulence.

Spray coverage profiles from pecan air-blast sprayers, with a radial air-flow and a volute-generated focused air-flow, as affected by forward speed and application volume

A portable scanning system for evaluation of spray deposit distribution