Ultrasonic sensing of pistachio canopy for low-volume precision spraying

doi:10.1016/j.compag.2014.12.015

Computers and Electronics in Agriculture

Volume 112, March 2015, Pages 149-160

https://doi.org/10.1016/j.compag.2014.12.015 Get rights and content

Highlights

•
Precision control system for the estimation of tree canopy dimensions was designed.
•
Three ultrasonic ranging sensors were utilized to estimate the target distance.
•
A MLP neural network was used for volume estimation of tree sections.
•
Consistency of spray distribution was represented in selective application.
•
Variable-rate spraying showed a reduction in pesticide usage of about 34.5%.

Abstract

Effective volume rate of pesticide application on a site-specific basis can reduce the amount of agrochemicals used in precision horticulture. The prototype sprayer used in this study provided volume application rate adapted to the canopy volume in pistachio orchards on a real-time and continuous basis. An electronic control system for the detection and estimation of tree canopy dimensions was designed for application rate adjustment. Three ultrasonic ranging USS3 sensors were utilized to estimate the distance to the target at three different heights. A MLP neural network with gradient-descent back-propagation algorithm, tangent-sigmoid transfer function, and 3-7-6 topology was used for volume estimation of tree sections. Training and validation errors as well as R² values indicated the reliability of the network for volume prediction. Results of T-test for comparing the number of spray droplet impacts, coverage of (artificial) target, spray quality parameter and relative span factor between variable-rate and conventional spraying were not significant which indicates the consistency of spray distribution in selective application. Experiments showed a reduction in pesticide usage of about 34.5% by means of variable-rate technology (41.3, 25.6 and 36.5, respectively for the top, middle, and bottom sections of tree canopy). Precise application of agrochemicals reduces both costs and environmental pollution by supporting a decrease in the amount of delivered spray.

Introduction

Variable rate spraying of the canopy allows growers to apply adjusted volume rate of pesticides to the target, based on canopy size, season and growth stage and to apply plant protection products in an economic and environmentally sound manner. Since four decades ago, ultrasonic sensors have been employed in agriculture for different purposes (Planas et al., 2011). One of these applications is detection and ranging to obtain structural data from trees. The first advances in this field were related to the application of plant protection materials such as pesticides in different orchards. When dose adjustment according to canopy structure was proposed (Byers et al., 1971, Morgan, 1964) some researchers began to design electronic systems for measuring canopy structural parameters. The first proposed systems to determine canopy volume used many ultrasonic sensors on a vertical mast (McConnell et al., 1983) or mounted on the sprayer (Giles et al., 1988). Because of the state-of-the-art of the application technologies, using this information in real time was not possible. Usage of ultrasonic sensors has been reported only for detection of canopy presences by some researchers (Brown et al., 2008, Giles et al., 1987). In this method spraying was done exclusively when canopy was in front of the sprayer. Another application was citrus trees spraying from constant given distance (Moltó et al., 2000). The nozzles are located on a movable arm which follows the boundary of the tree according to data collected from sensors. In this research ultrasonic sensors were placed 50 and 75 cm apart. The same authors improved another sprayer which was able to spray with 3 different dosage according to width estimation of canopy made by ultrasonic sensors (Moltó et al., 2001). There was 1-without spraying when there was no vegetation, 2-half spraying when there was little vegetation in front of the sensors and 3-full spraying when sensors detected width of canopy above a given threshold. This achievement led the way to a continues variation of flow rate according to variability of the canopy along citrus groves, vineyard and fruit orchards rows (Escolà et al., 2002, Escolà et al., 2007, Escolà et al., 2013, Gil et al., 2007, Llorens et al., 2010, Solanelles et al., 2002, Solanelles et al., 2006). Different researches have been conducted for automatic measurement of canopy dimensions in citrus groves. Early works were concentrated on comparing of manual volume estimation with LIDAR and ultrasonic sensors measurements (Tumbo et al., 2002). Results indicated good correlation between the estimation made by LIDAR and ultrasonic sensors, while correlation with manual measurements was lower. Differences between electronic and manual measurements were attributed to higher resolution of data gathered with sensors by authors. This system included a vertical mast with equipped 10 ultrasonic sensors. Alternated sensors were fired in different groups sequentially, to prevent signal interference. This system was later implemented for providing a canopy volume map in citrus groves with attached DGPS receiver (Schumann and Zaman, 2005). Observation showed larger differences between manual and sensor estimations in less dense trees. This canopy information was used to adjust fertilizer dose rate (Zaman et al., 2005) and estimate fruit yield in citrus groves (Zaman et al., 2006). LIDAR sensor in relation to vertical sampling resolution can gather much more information from canopy parameters for a more accurate estimation in comparison with array of ultrasonic sensors (Gil et al., 2013, Palleja et al., 2010, Rosell et al., 2009). The results of these tests were satisfactory, but extrapolation of these results to trees with different structures is not easy.

Although several groups have developed prototypes to adjust the application flow rate to the variations in canopy structural parameters using ultrasonic sensors, a review of various targeted spraying methods (Van de Zande et al., 2008) showed that solutions for variable rate spraying in orchards are still in prototype phase, however, there are already commercially available sprayers for weed control and plant fertilization in arable land. The objective of this research was to develop a prototype that can apply a variable amount of pesticide according to the canopy variability along the crop row for agrochemical applications in pistachio orchards. Thus this paper consist of two main parts: (a) a description of the electronic system for measurements of canopy structural parameters and the calculation of the adapted flow rate; and (b) comparing of the prototype with conventional spraying to evaluate benefits of variable application method. The goal of this study is improving computational method for calculating of the output flow rate using artificial neural network, in order to solve the problem there has been in the previous researches. By automatically spraying the optimal amount of spray mixtures into tree canopies and seizing the spray application beyond target areas, the variable rate sprayer with automatic control can reduce the amount and cost of pesticides for growers, reduce the risk of environmental pollution by pesticides, and provide safer and healthier working conditions for workers.

Section snippets

Materials and methods

The variable rate prototype for precision spraying and its components are presented in this section (Fig. 1). This prototype consists of two main mechanical and electronic parts. Data collected from the sensors and the shaft encoder are delivered to the microcontroller for data processing and the output signal is sent to the actuator in order to adjust valve openings. First, changes made in the configuration of the conventional sprayer are mentioned and then the electronic parts will be

Evaluation methodology

Two different treatments were established in the pistachio orchard. The experiments were carried out in the Peek Village near the city of Saveh (Latitude 50′ 44″ and longitude 35′ 17″). In the first treatment, a conventional sprayer was calibrated to apply a constant rate of 810 l ha⁻¹, based on the results obtained in preliminary tests. In the second treatment, pesticide flow rate was continuously varied according to crop structure using the ANN method of dose adjustment.

Row spacing was 4 m and

Results and discussion

Different Artificial neural networks with various training algorithms and transfer functions were tested to obtain the best network for prediction of canopy volume at different tree sections (V₁, V₂, V₃). Based on mean square error and R² criteria, the best ANN model selected (Fig. 13). After much trial and error tests, the best ANN was obtained with three layers (6 inputs, 7 hidden nodes and 3 outputs) having MSE of 0.0026 (Fig. 14) and R² of 0.9923 for training and MSE of 0.012 and R² of

Conclusions

A canopy-sensing, automatically-controlled sprayer was successfully developed and operated in a pistachio orchard. The system utilized ultrasonic sensors to determine the variation in canopy structure and, based on calculation, adjust valve opening to implement variable-rate application. The system can be used in sensor-based precision agriculture for implementing site-specific management in variable-rate application of chemical inputs. Comparative tests between the treatments obtained from the

References (36)

D. Brown et al.
Targeted spray technology to reduce pesticide in runoff from dormant orchards
Crop Prot.
(2008)
J. Cross et al.
Spray deposits and losses in different sized apple trees from an axial fan orchard sprayer: 1. Effects of spray liquid flow rate
Crop Prot.
(2001)
A. Escolà et al.
Variable rate sprayer. Part 1 – Orchard prototype: design, implementation and validation
Comput. Electron. Agric.
(2013)
E. Gil et al.
Variable rate application of plant protection products in vineyard using ultrasonic sensors
Crop Prot.
(2007)
E. Gil et al.
Variable rate sprayer. Part 2 – Vineyard prototype: design, implementation, and validation
Comput. Electron. Agric.
(2013)
J. Llorens et al.
Variable rate dosing in precision viticulture: use of electronic devices to improve application efficiency
Crop Prot.
(2010)
E. Moltó et al.
PM—power and machinery: design and testing of an automatic machine for spraying at a constant distance from the tree canopy
J. Agric. Eng. Res.
(2000)
E. Moltó et al.
PM—power and machinery: pesticide loss reduction by automatic adaptation of spraying on globular trees
J. Agric. Eng. Res.
(2001)
T. Palleja et al.
Sensitivity of tree volume measurement to trajectory errors from a terrestrial LIDAR scanner
Agric. For. Meteorol.
(2010)
G. Pergher et al.
Foliar deposition and pesticide losses from three air-assisted sprayers in a hedgerow vineyard
Crop Prot.
(1997)

A. Schumann et al.

Software development for real-time ultrasonic mapping of tree canopy size

Comput. Electron. Agric.

(2005)

F. Solanelles et al.

An electronic control system for pesticide application proportional to the canopy width of tree crops

Biosyst. Eng.

(2006)

L. Albrigo et al.

Yield estimation of ‘Valencia’ orange research plots and groves

Proc. Fla. State Hortic. Soc.

(1975)

R. Byers et al.

Base gallonage per acre

Virginia Fruit

(1971)

Chen, Y., 2010. Development of an intelligent sprayer to optimize pesticide applications in nurseries and orchards...

Escolà, A., Solanelles, F., Planas, S., Rosell, J.R., 2002. Electronic control system for proportional spray...

A. Escolà et al.

Variable dose rate sprayer prototype for tree crops based on sensor measured canopy characteristics

D. Giles et al.

Control of orchard spraying based on electronic sensing of target characteristics

Trans. ASAE

(1987)

Cited by (78)

Economic and environmental benefits of digital agricultural technologies in crop production: A review
2024, Smart Agricultural Technology
This comprehensive review delved into the economic and environmental benefits of Digital Agricultural Technologies (DATs) in crop production, synthesising data from 136 peer-reviewed papers and 28 documents with empirical data from relevant EU projects. This analysis highlighted the substantial contribution of DATs across five key categories: Recording and Mapping Technologies (RMT), Guidance and Controlled Traffic Farming (CTF) Technologies, Variable Rate Technologies (VRT), Robotic Systems or Smart Machines (RSSM), and Farm Management Information Systems (FMIS). Specifically, it provided an overview of the various benefits that these technologies can deliver with the most significant ones revealing reductions of up to 80 % in fertiliser usage with RMT and CTF applications, while VRT demonstrated a 60 % decrease in fertiliser usage and up to 80 % reduction in pesticide use. VRT also showed an increase in yield by 62 %. RSSM was able to reduce labour by 97 % and diesel consumption by 50 %. FMIS improved yield by 10 % to 15 %, facilitating simultaneous reductions in labour and input costs, illustrating the critical role of integrated digital solutions in enhancing agricultural efficiency and sustainability. From an environmental point of view, VRT has emerged as a major factor in environmental sustainability, demonstrating water savings of 20 % to 50 % in vineyards and pear orchards and a significant reduction in greenhouse gas emissions. These findings highlighted the significant benefits of DATs on enhancing productivity and promoting environmental sustainability. They provided a compelling case for further investment and research in DATs through quantifiable benefits in crop production.
Development of a line following autonomous spraying vehicle with Machine vision-based leaf density estimation for cherry tomato greenhouses
2023, Computers and Electronics in Agriculture
Traditional manual spraying systems and even modern automated ones often neglect the critical factors of plant growth stage and coverage density during spraying for plants, which can result in suboptimal yields. To address this issue, this study aims to describe the development of an efficient and affordable autonomous spraying vehicle (ASV) for application in cherry tomato greenhouses. The ASV integrates a visual autonomous spraying system (VASS) and a vehicle control system (VCS). The VASS uses leaf density signalling with machine vision and a vertical spray boom to target leaves for precision spraying application. Leaf density signalling is an algorithm developed to recognise and estimate leaf density by machine, simplifying the real-time determination of leaf density levels in plants. The performance of the algorithm demonstrates an accuracy of 84 % with an average video processing speed of 33.3 ms. The VASS is integrated into VCS to facilitate remote or autonomous navigation. The visual line-following system (VLFS) supports ASV in autonomously navigating between narrow-spaced rows in cherry tomato greenhouses, thereby automating the entire spraying process. The analysis of the comprehensive performance of the ASV and the comparative analysis with knapsack sprayers in field experiments are discussed and presented. Comparative analysis with manually operated sprayers demonstrates the better performance of the ASV. These experiments highlight the potential of the ASV to replace traditional manual spraying methods while enhancing spraying precision.
Modelling wine grapevines for autonomous robotic cane pruning
2023, Biosystems Engineering
Aotearoa (New Zealand) has a strong and growing winegrape industry struggling to access workers to complete skilled, seasonal tasks such as pruning. Maintaining high-producing vines requires training agricultural workers that can make quality cane pruning decisions, which can be difficult when workers are not readily available. A novel vision system for an autonomous cane pruning robot is presented that can assess a vine to make quality pruning decisions like an expert. The vision system is designed to generate an accurate digital 3D model of a vine with skeletonised cane structures to estimate key pruning metrics for each cane. The presented approach has been extensively evaluated in a real-world vineyard as a commercial platform would be expected to operate. The system is demonstrated to perform consistently at extracting dimensionally accurate digital models of the vines. Detailed evaluation of the digital models shows that 51.45% of the canes were modelled entirely, with a further 35.51% only missing a single internode connection. The quantified results demonstrate that the robotic platform can generate dimensionally accurate metrics of the canes for future decision-making and automation of pruning.
Construction and testing of a model for predicting droplet deposition based on Gaussian process regression
2023, Computers and Electronics in Agriculture
To realize precision spraying, predicting the pesticide application dose based on the structural characteristics of the plant canopy is a crucial step. However, most of the current precision spraying techniques are based on canopy characteristics with no consideration given to the effect of canopy characteristics on droplet particle population. As a result, the droplet deposition value is far from being ideal. Herein, we explored the effects of canopy density and spraying volume on the distribution of droplet deposition in simulated citrus tree canopies and constructed a prediction model for droplet deposition. In this research, the three-dimensional (3D) reconstruction of fruit tree canopy was obtained using binocular camera and a new method was proposed to describe simulated citrus tree canopy density using three-dimensional point cloud. The optimal description of the interrelationship between 3D point cloud and fruit tree canopy density was determined by comparing the determination coefficients, Akaike information criterion and relative errors of the four fitting functions. Through our experiments, a droplet deposition prediction model based on Gaussian process regression was established by changing the duty cycle of the solenoid valve to adjust the spray volume, with simulated citrus trees as the target and water sensitive paper as the sampling method for collecting droplet deposition data. The experimental results show that the distribution of droplet deposition in the upper and lower layers of the canopy varies, but within the same layer droplets can produce relatively uniform deposition in similar canopy density areas with correlation coefficients greater than 0.6. When the canopy density reached 4.35 m²/m³, the amount of droplet deposition was anomalous. The prediction model, after training, has good ability of predicting droplet deposition with R² higher than 0.79. At the same time, it was found in the research that differences occurred between the prediction accuracy of the droplet deposition prediction model in the upper canopy layer and that in the lower canopy layers, in which the R² of the prediction model in the lower layer could reach 0.82, while in the upper layer it was 0.79. It is expected that our research on droplet deposition prediction can provide a new research perspective for precision spraying technology in plant protection operations.
Evaluation of an electrostatic spray charge system implemented in three conventional orchard sprayers used on a commercial apple trees plantation
2023, Crop Protection
Electrostatic sprayers can be an option for increasing the effectiveness of treatments and reducing pesticide consumption, but it still requires more verifications, especially in orchards. The spray deposition performances of three different electrostatic sprayers design were compared for apple tree pesticide application, all of them used with the electrostatic system activated and disactivated, as conventional spray application. Despite designs of particular type sprayers were unique, no significant differences were observed in the normalized deposition and droplets density between the conventional and electrostatic spray methods of the same sprayer, but differences were found between the vertical boom and the pneumatic sprayer in terms of coverage. The normalized deposition values of the three sprayers were very similar, ranging between 1.0 and 1.5 μg cm⁻². However, the vertical boom sprayer produced the highest spray coverage, followed by the multi-fan sprayer. In general, the measured values of coverage and deposit obtained were in the three cases similar, independently on the use or not of the electrostatic device, being the sprayer type more affecting parameter rather than the electrostatic charging system. Differences in spray deposition qualities on apple trees were observed among three types of sprayers, but were not for the same sprayer with and without activation of the electrostatic charging system. Appropriate design of orchard sprayers according to the canopy structure seems to be more effective rather than the implementation of electrostatic spray charge system to achieve the European challenge of 50% reduction of pesticides by 2030.
Design, implementation and validation of a sensor-based precise airblast sprayer to improve pesticide applications in orchards
2024, Precision Agriculture

View all citing articles on Scopus

View full text

Ultrasonic sensing of pistachio canopy for low-volume precision spraying

Highlights

Abstract

Introduction

Section snippets

Materials and methods

Evaluation methodology

Results and discussion

Conclusions

Crop Prot.

Crop Prot.

Comput. Electron. Agric.

Crop Prot.

Comput. Electron. Agric.

Crop Prot.

J. Agric. Eng. Res.

J. Agric. Eng. Res.

Agric. For. Meteorol.

Crop Prot.

Comput. Electron. Agric.

Biosyst. Eng.

Yield estimation of ‘Valencia’ orange research plots and groves

Proc. Fla. State Hortic. Soc.

Base gallonage per acre

Virginia Fruit

Variable dose rate sprayer prototype for tree crops based on sensor measured canopy characteristics

Control of orchard spraying based on electronic sensing of target characteristics

Trans. ASAE