Biomass production is a diagnosis tool for the evaluation of the effect of climate, crop genomic and management. The differences in biomass accumulation are necessary for the assessment of the fertilization necessities in the strategies for variable nitrogen doses. Remote sensing-based data provide a direct observation of the differences in canopy development across time and space and can be integrated into the physiological basis of crop growth models to provide estimates of biomass production at fine scales. The proposed approach was applied in a wheat field in Albacete, Spain and the results were compared with measurements of aboveground biomass and yield maps obtained by a combined-mounted grain yield monitor.

In this study, we investigated the potential of airborne imaging spectroscopy for in-season grassland yield estimation. We utilized an unmanned aerial vehicle and a hyperspectral imager to measure radiation, ranging from 455 to 780 nm. Initially, we assessed the spectral signature of five typical grassland species by principal component analysis, and identified a distinct reflectance difference, especially between the erectophil grasses and the planophil clover leaves. Then, we analyzed the reflectance of a typical Norwegian sward composition at different harvest dates. In order to estimate yields (dry matter, DM), several powered partial least squares (PPLS) regression and linear regression (LR) models were fitted to the reflectance data and prediction performance of these models were compared with that of simple LR models, based on selected vegetation indices and plant height. We achieved the highest prediction accuracies by means of PPLS, with relative errors of prediction from 9.1 to 11.8% (329 to 487 kg DM ha−1) for the individual harvest dates and 14.3% (558 kg DM ha−1) for a generalized model.

Grassland silage management is generally ad hoc resulting in soil compaction damage. Literature suggests grass yield reductions of 5 to 74% through compaction (UK mean 13%), while a 2015 study, reported here, comparing grass dry matter (DM) yield between controlled traffic farming (CTF) and normal management (N), found a 13.5% (0.80 t ha−1) increase for CTF. Commercially available grass forage equipment with widths of 3 to 12 m set up for CTF reduced trafficked areas from 80%–90% for N to 40%–13%. Economic analysis based on 13% increase in DM for 2 and 3 cut systems, gave an increased grass value between £38 ha−1 and £98 ha−1. CTF for multi-cut grass silage effectively increases yields by reducing compaction and sward damage.

Great technological advances have been made in Precision Agriculture (PA) in the past decade, yet adoption of PA in intensive grassland areas in North West Europe is low. This is despite the fact that in these areas the market structures are suitable and there are highly developed agricultural and food industries offering great potential for the application of new technology. Specific inefficiencies in plant nutrient management in soil exist, which are not only limiting grass yields but are also causing environmental deterioration. Soil nutrient management efficiency could be greatly improved using PA techniques, but the complexity of grassland systems, coupled with a lack of calibration of sensors specific to grassland, together with local barriers, appear to be the reasons why PA adoption is poor in these areas. This paper reviews new and existing technology including soil and crop sensors, navigation devices, remote sensing and unmanned aerial vehicles. The suitability and readiness of these technologies for adoption in grassland areas is discussed, along with data interpretation issues, future perspectives and research opportunities.

Yield and landscape are commonly used to guide management zone delineation. However, production system choice and management can interact with landscape attributes and weather. The objective of this study was to evaluate forage yield and soil properties in three landscape defined (elevation based) management zones, and under two different grazing systems. Changes in soil properties (soil strength, bulk density, moisture, bioavailable nutrients) and forage productivity (biomass), as related to grazing management and management zone, were measured. Bulk density, moisture, and forage biomass were greater at higher elevation. Soil strength decreased as elevation increased, and was greater near-surface after winter grazing ended. The response of landscape delineated management zones varied with extreme weather conditions and treatment. Lower zones were more sensitive to weather extremes than higher elevations, directly affecting biomass accumulation. In conclusion, we observed interactions between the grazing treatments and the management zones.

This study aims to evaluate actual biomass and N-uptake estimates with the Yara N-Sensor in intensively managed grass swards across several trial sites in Europe. The dataset was split by location into an independent calibration data (UK and Finland) and a validation data (Germany) for the first two cuts. Yara N-Sensor readings were better correlated with N-uptake (R2=0.71) than actual biomass (R2=0.53) for the 1st cut. At the 2nd cut, the R2 values for both parameters were higher (0.80 and 0.56). A cross-validation with a German grass trial indicated the potential for predicting N-uptake (R2>0.8). It can be concluded that the technology has the potential to guide management decisions and variable rate nitrogen application on European grass swards.

The main objective of this work was to evaluate technologies that have potential for monitoring aspects related to spatial and temporal variability of soil nutrients and pasture yield and for support to decision making by the farmers. Three types of sensors were evaluated: an electromagnetic induction sensor, an active optical sensor and a capacitance probe. The results are relevant for the selection of the adequate sensing system for each particular application and to open new perspectives for other works that would allow the testing, calibration and validation of the sensors in a wider range of pasture production conditions and rainfall patterns, characteristic of the Mediterranean region.

As the interest in Unmanned Aerial Systems (UAS) has increased, so has the interest in the application of these systems for use in agriculture. A variety of sensors, including Multi-Spectral, Near-Infrared, Thermal, and True-Color have the potential to benefit farmers when mounted to a UAS. But as this is an emerging field, there is little data available to demonstrate their use for early detection of plant diseases in crop production. In 2016, a preliminary study was launched to examine the potential of using aerial imagery from UAS to detect diseases in soybean crops. Two irrigated fields in Alabama were selected: Experiment 1, a 50-hectare field, and Experiment 2, a 5-hectare field. Each trial consisted of replicated plots using two foliar fungicide treatments and an untreated control. Aerial imagery (multi-spectral and true-color) was collected on a biweekly basis during this study. Using multi-spectral imagery, both the Normalized Difference Vegetative Index (NDVI) and Normalized Difference Red Edge Index (NDRE) were generated and compared to direct observations in the field. Disease severity of soybean rust, charcoal rot and Cercospora leaf blight were monitored on a biweekly basis and correlated to the UAS imagery. Preliminary results indicated plant stress can be detected using UAS imagery. In Experiment 1, stress associated with charcoal rot was visible in the NDRE imagery. This was of interest because at the time of flight, while it was noted that plants were yellowing, the root and stem disease itself had not been identified by direct observation. In Experiment 2, soybean rust was observed by direct observation and in both the NDRE and NDVI imagery. Soybean rust did have a negative impact on yield in Experiment 2, however severe drought conditions may have negated the yield loss likely caused by the development of charcoal rot in Experiment 1.

For most producers, unmanned aerial vehicles (UAV) are a novelty that has been little employed in their agricultural operations. An UAV will not fix every problem on the farm, but there are some practical applications for which UAVs have demonstrated value. Three examples of how UAVs have been used in weed science applications are presented here; the methods are transferable to other agricultural commodities with similar characteristics. The first of these is quantification of the extent and severity of non-target herbicide injury. The second application is calculation of spray thresholds based on weed populations. The third application is development of site-specific herbicide treatment.

This paper investigates the effectiveness of using a UAV with dual commercial off-the-shelf (COTS) cameras, one un-modified and one modified to sense near infra-red (NIR) wavelengths to identify the onset of disease within a trial crop of potatoes. The trial was composed of 2 plots of 16 drills containing 12 tubers exposed to the blackleg disease-causing bacterial pathogen (Pectobacterium atrosepticum) in order to demonstrate best practise tuber storage and haulm destruction methods. Eleven sets of aerial data were gathered between 27/5/2016~29/7/2016 and compared with ground truth data collected on 14/7/2016. Visual analysis of the data could only detect the onset of disease and not the specific infection and resulted in a user accuracy (UA) of 83% and producer accuracy (PA) of 78%, with a total accuracy (TA) of 91% and Kappa coefficient (K) of 0.75. The building blocks of an automated classification routine have been constructed using pixel and object based image analysis (OBIA) methods, which have shown promising first results (UA 65%, PA 73%, TA 87%, K 0.61) but requires further refinement to achieve an equivalent level of accuracy as that of the visual analysis.

New remote sensing technologies have provided unprecedented results in vineyard monitoring. The aim of this work was to evaluate different sources of images and processing methodologies to describe spatial variability of spectral-based and canopy-based vegetation indices within a vineyard, and their relationship with productive and qualitative vine parameters. Comparison between image-derived indices from Sentinel 2 NDVI, unfiltered and filtered UAV NDVI and with agronomic features have been performed. UAV images allow calculating new non-spectral indices based on canopy architecture that provide additional and useful information to the growers with regards to within-vineyard management zone delineation.

By providing high spatial resolution images of vine fields througout the vine growing season, UAVs could provide useful information, different than those normally considered in the literature. This study aimed at identifying i) relevant information that can be observed from UAV images by two kind of stakeholders : growers and advisers (G&A) ii) the most suitable periods for this observation iii) and the added value this information can have for both G&A daily tasks. This approach has been conducted on an 11.3 ha commercial vineyard representative of the south of France vineyards. UAV-based visible images (2.5 cm resolution) have been acquired in commercial conditions every two weeks from budburst to harvest. Images have been presented to two groups of G&A six times during the growing season. Every expert gathering was conducted with i) an individual period where images were presented to each expert ii) a collective period where G&As were invited to share information. Application of this methodology demonstrated that much information about the vines, the soil and the vineyard environment can be extracted from UAV-based visible images without any image processing. Results showed that this information have added value all along the growing cycle of the vine, particularly for advisers.

Precision agriculture (PA) requires reasonably homogeneous areas for site-specific management. This work explores the applicability of digital terrain classes obtained from a digital elevation model derived from UAV-acquired images, to define management units in in a relative flat area of about 6 ha. Elevation, together with other terrain variables such as: slope degree, profile curvature, plan curvature, topographic wetness index, sediment transport index, were clustered using the Fuzzy Kohonen Clustering Network (FKCN). Four terrain classes were obtained. The result was compared with a map produced by a classification of soil properties previously interpolated by ordinary kriging. The results suggest that areas for site-specific management can be defined from terrain classes based on environmental covariates, saving time and cost in comparison with interpolation of soil variables.

In this study, the biomass of winter wheat was estimated by using hyperspectral data obtained from a hyperspectral camera on an Unmanned Aerial Vehicle (UAV). Every two bands from the hyperspectral data were selected to calculate two kinds of vegetation indexes: the Normalized Difference Vegetation Index (NDVI) and Ratio Vegetation Index (RVI). Linear models were established between winter wheat biomass and those indexes, and coefficient of determination R2 was used to draw the two-dimensional distribution of R2 values. The comparison between NDVI and RVI for pixel covered by soil and wheat showed that RVI is more efficient to mask the influence from soil than NDVI. For calculating the NDVI, optimal bands are located mainly around 820 nm and 725 nm to 750 nm. For assessing RVI, the wavelength range from 820 to 832 nm, 794 to 808 nm, 770 to 788 nm, 725 nm to 750 nm and 890 nm for RVI are most suitable. Those optimal bands can achieve a coefficient of determination R2 higher than 0.88 by using the linear regression model in the study.

This paper reports an investigation of the relationship between spray characteristics and a nozzles’ internal structure to reveal the working mechanism of anti-drift spray nozzles. Three important structural factors were taken into account, the diameter of the inner chamber, the angle of V-shaped slot and the relative kerf depth. Three-dimensional models of the fan nozzles were set up using Solidworks software and the corresponding real nozzles were produced using high-precision 3-D printer. The flow fields inside the nozzles were simulated using the software FLUENT. By comparing the flow fields inside and outside the nozzles under the conditions of the same inner structural parameter, the relationships between spraying flow characteristics and different structural parameters was made clear, and provides a reference for optimal design of anti-drift spray nozzles.

This paper presents a method for automating weed detection in colour images despite heavy leaf occlusion. A fully convolutional neural network is used to detect the weeds. The network is trained and validated on a total of more than 17,000 annotations of weeds in images from winter wheat fields, which have been collected using a camera mounted on an all-terrain vehicle. Hereby, the network is able to automatically detect single weed instances in cereal fields despite heavy leaf occlusion.

Precise applying of PPP (Plant Protection Products) in orchards and vineyards requires new kinds of sprayer technologies and new methods of sensor data evaluation. In this paper a selective electrical driven sprayer, carried by the autonomous robotic platform elWObot, is introduced. A 3D-Simulation environment and the framework ROS (Robot Operating System) helps developing and testing the interaction between the sprayer and the robot. The calculated leaf wall area (LWA) and the distance from the sprayer to the leaves in the spray region, control the flow-rate and the air-assist of eight adjustable sprayers individually. First field trials showed that the adaption of the software from the simulation to the hardware worked as expected.

The difficulties of transporting heavy mobile robots limit robotic experiments in agriculture. Virtual reality however, offers an alternative to conduct experiments in agriculture. This paper presents an application of virtual reality in a robot navigational experiment using SolidWorks and simulated into MATLAB. Trajectories were initiated using Probabilistic Roadmap and compared based on travel time, distance and tracking error, and the efficiency was calculated. The simulation results showed that the proposed method was able to conduct the navigational experiment inside the virtual environment. U-turn trajectory was chosen as the best trajectory for crop inspection with 82.7% efficiency.

In order to exploit potentials of 20–40% reduction of herbicide use, as documented by use of Decision Support Systems (DSS), where requirements for manual field inspection constitute a major obstacle, large numbers of digital pictures of weed infestations have been collected and analysed manually by crop advisors. Results were transferred to: 1) DSS, which determined needs for control and connected, optimized options for control returned options for control and 2) convolutional, neural networks, which in this way were trained to enable automatic analysis of future pictures, which support both field- and site-specific integrated weed management.

The purpose of this work was to evaluate the capacity of existing landscape irrigation equipment to deliver the expected irrigation depth at a level of uniformity that can potentially be used in a precision irrigation management environment. Popular pop-up sprayer systems from two leading manufacturers (Rain-bird and Hunter variable arc and fixed arc nozzles) were subject to spatial distribution tests, in order to establish the application patterns under ideal conditions. The results indicate that it is not possible to fully rely on the catalog application depth values, with some variable arc nozzles applying up to 200% of the catalog values. These nozzles can have low values of uniformity, with Christiansen´s uniformity coefficient, CU, of between 0.31 and 0.70, making it difficult to obtain good uniformity even under a back-to-back setting. These results indicate that variable arc nozzles are not recommended for use in precision agriculture. It is preferable to use fixed arc nozzles, after actual field evaluation.