Site-specific weed management (on the scale of a few meters or less) has the potential to greatly reduce pesticide use and its associated environmental and economic costs. A prerequisite for site-specific weed management is the availability of accurate maps of the weed population that can be generated quickly and cheaply. Improvements and cost reductions in unmanned aerial vehicles (UAVs) and camera technology mean these tools are now readily available for agricultural use. We used UAVs to collect aerial images captured in both RGB and multispectral formats of 12 cereal fields (wheat [Triticum aestivum L.] and barley [Hordeum vulgare L.]) across eastern England. These data were used to train machine learning models to generate prediction maps of locations of black-grass (Alopecurus myosuroides Huds.), a prolific weed in UK cereal fields. We tested machine learning and data set resampling methods to obtain the most accurate system for predicting the presence and absence of weeds in new out-of-sample fields. The accuracy of the system in predicting the absence of A. myosuroides is 69% and its presence above 5 g in weight with 77% accuracy in new out-of-sample fields. This system generates prediction maps that can be used by either agricultural machinery or autonomous robotic platforms for precision weed management. Improvements to the accuracy can be made by increasing the number of fields and samples in the data set and the length of time over which data are collected to gather data across the entire growing season.

Controlling weeds is a critically important task in sugarcane production systems. Weeds compete for light, nutrients, and water, and if they are not managed properly can negatively impact sugarcane yields. Accurate detection of weeds versus desired plants was assessed using hyperspectral and pigment analyses. Leaf samples were collected from four commercial Louisiana sugarcane varieties, and nine weed species commonly found in sugarcane fields. Hyperspectral leaf reflectance data (350 to 850 nm) were collected from all samples. Plant pigment (chlorophylls and carotenoids) levels were also determined using high-performance liquid chromatography, and concentrations were determined using authentic standards and leaf area. In all cases, leaf reflectance data successfully differentiated sugarcane from weeds using canonical discrimination analysis. Linear discriminant analysis showed that the accuracy of the classification varied from 67% to 100% for individual sugarcane varieties and weed species. In all cases, sugarcane was not misclassified as a weed. Plant pigment levels exhibited marked differences between sugarcane varieties and weed species with differences in chlorophyll and carotenoid explaining much of the observed variation in reflectance. The ratio of chlorophyll a to chlorophyll b showed significant differences between sugarcane and all weed species. The successful implementation of this technology as either an airborne system to scout and map weeds or a tractor-based system to identify and spray weeds in real-time would offer sugarcane growers a valuable tool for managing their crops. By accurately targeting weeds in sugarcane fields that are emerged and growing, the total amount of herbicide applied could be decreased, resulting in cost savings for the grower and reduced environmental impacts.

Thanks to recent developments in service robotics technologies, precision agriculture (PA) is becoming an increasingly prominent research field, and several studies were made to present and outline how the use of mobile robotic systems can help and improve farm production. In this paper, the integration of a custom-designed mobile base with a commercial robotic arm is presented, showing the functionality and features of the overall system for crop monitoring and sampling. To this aim, the motion planning problem is addressed, developing a tailored algorithm based on the so-called manipulability index, that treats the base and robotic arm mobility as two independent degrees of motion; also developing an open source closed-form inverse kinematics algorithm for the kinematically redundant manipulator. The presented methods and sub-system, even though strictly related to a specific mobile manipulator system, can be adapted not only to PA applications where a mobile manipulator is involved but also to the wider field of assistive robotics.

Timely estimation of silage maize protein provides an effective decision to adapt optimized strategies for nitrogen fertilizer management and also harvesting time for farmers. So, this research aimed to investigate whether using vegetative indices (VIs) derived from UAV remotely sensed multispectral (with 520–900 nm wavelengths) imagery and also Soil Plant Analysis Development (SPAD) greenness index can be used to detect the leaf protein concentration (LPC) of silage maize, as a function of various nitrogen rates (0, 50, 100, and 150% of recommended dosage). Results of principal component analysis (PCA) suggested that LPC was highly correlated with leaf greenness index in all developmental stages. In addition, LPC was highly correlated with most of the VIs investigated. A PCA clustering showed the meaningful pattern of N rates. Higher LPC values, VIs, and greenness index were more concentrated in the higher nitrogen (N100% and N150%) sectors. Nitrogen Reflectance Index (NRI) was identified as the most important VIs to monitor and predict LPC in the silage maize field, showing a strong polynomial relationship with LPC in both eight-leaf collar (V8) (R2 = 0.81, p ≤ 0.01) and tasseling (VT) (R2 = 0.98, p ≤ 0.001) stages. In addition, among VIs, the Normalized Difference Vegetation Index (NDVI) demonstrated a significant linear regression relationship with LPC (R2 = 0.80, p ≤ 0.01) in the VT. Findings suggested the high potential of VIs extracted by UAV-taken multispectral imagery and also SPAD proximal sensing to help farmers rapidly diagnose LPC in silage maize, in line with the objectives of precision farming.

The past 50 yr of advances in weed recognition technologies have poised site-specific weed control (SSWC) on the cusp of requisite performance for large-scale production systems. The technology offers improved management of diverse weed morphology over highly variable background environments. SSWC enables the use of nonselective weed control options, such as lasers and electrical weeding, as feasible in-crop selective alternatives to herbicides by targeting individual weeds. This review looks at the progress made over this half-century of research and its implications for future weed recognition and control efforts; summarizing advances in computer vision techniques and the most recent deep convolutional neural network (CNN) approaches to weed recognition. The first use of CNNs for plant identification in 2015 began an era of rapid improvement in algorithm performance on larger and more diverse datasets. These performance gains and subsequent research have shown that the variability of large-scale cropping systems is best managed by deep learning for in-crop weed recognition. The benefits of deep learning and improved accessibility to open-source software and hardware tools has been evident in the adoption of these tools by weed researchers and the increased popularity of CNN-based weed recognition research. The field of machine learning holds substantial promise for weed control, especially the implementation of truly integrated weed management strategies. Whereas previous approaches sought to reduce environmental variability or manage it with advanced algorithms, research in deep learning architectures suggests that large-scale, multi-modal approaches are the future for weed recognition.

Pesticide regulations and application technologies are changing rapidly due to rising concerns around off-target movement of pesticides and increased focus on improving the efficiency of pesticide applications. In order to conduct relevant applied research and develop educational programs related to pesticide application, it is necessary to understand the common application practices and technologies that growers use. A survey was conducted to assess common pesticide application practices and technologies used by Georgia growers. Both online and printed survey copies were distributed by county agricultural extension agents to growers in all 159 counties. A total of 186 responses representing agronomic crops in 65 counties were received and analyzed for results. Main results of this survey indicated that 1) 72% of respondents produced ≥200 ha of crops; 2) 29% of respondents received their information from university Extension personnel; 3) 42% of respondents used a separate sprayer for applications of dicamba, 2,4-D, or 2,4-DB; 4) 46% of respondents used sprayers with boom lengths ≥18.3 m; 5) 65% of respondents used ≥121 L/ha to apply pesticides; 6) 53% of respondents used three or more different nozzles on their spray booms throughout the season; 7) 68% of respondents used TeeJet® nozzles; 8) 65% of respondents used global positioning systems and rate controllers on their application equipment; 9) 66% of respondents recorded their pesticide application data on a notepad or diary; and 10) 39% of respondents reported that application accuracy is the biggest advantage of new spray technologies. Respondents also reported that weather, timing, and pesticide drift/regulations were their biggest application challenges and that more research is needed on topics such as rates, carrier volumes, pest control, chemicals and adjuvants. Information from this survey provides useful insights into the current application practices, technologies, and research needs of Georgia growers and will be used for developing appropriate research and educational efforts.

Numerous technologies have contributed to the recent development of agriculture, especially the advancement in hyperspectral remote sensing (HRS) constituted a revolution in crop monitoring. The widespread use of HRS to obtain crop parameters suggests the need for a review of research advances in this area. HRS offers new theories and methods for studying crop parameters, but much work needs to be done both experimentally and theoretically before we can truly understand the physical and chemical processes that predict these crop parameters. The study focuses on the following elements: 1) The article provides a relatively comprehensive introduction to HRS and how it can be applied to crop monitoring; 2) Current state-of-the-art techniques are summarized and analyzed to inform further advances in crop monitoring; 3) Opportunities and challenges for crop monitoring applications using HRS are discussed, and future research is summarized. Finally, through a comprehensive discussion and analysis, the article proposes new directions for using HRS to study crop characteristics, such as new data mining techniques including deep learning provide opportunities for efficient processing of large amounts of HRS data; combining the temporal and dynamic characteristics of crop parameters and vegetation growth processes will greatly improve the accuracy of crop parameter detection and monitoring; multidata fusion and multiscale data assimilation will become HRS monitoring. Multidata fusion and multiscale data assimilation will become another research hotspot for HRS monitoring of crop parameters.

Integrated pest management (IPM) seeks to minimize the environmental impact of pesticide application, and reduce risks to human and animal health. IPM is based on two important aspects – prevention and monitoring of diseases and insect pests – which today are being assisted by sensing and artificial-intelligence (AI) techniques. In this paper, we surveyed the detection and diagnosis, with AI, of diseases and insect pests, in cotton, which have been published between 2014 and 2021. This research is a systematic literature review. The results show that AI techniques were employed – mainly – in the context of (i) classification, (ii) image segmentation and (iii) feature extraction. The most used algorithms, in classification, were support vector machines, fuzzy inference, back-propagation neural-networks and recently, convolutional neural networks; in image segmentation, k-means was the most used; and, in feature extraction, histogram of oriented gradients, partial least-square regression, discrete wavelet transform and enhanced particle-swarm optimization were equally used. The most used sensing techniques were cameras, and field sensors such as temperature and humidity sensors. The most investigated insect pest was the whitefly, and the disease was root rot. Finally, this paper presents future works related to the use of AI and sensing techniques, to manage diseases and insect pests, in cotton; for instance, implement diagnostic, predictive and prescriptive models to know when and where the diseases and insect pests will attack and make strategies to control them.

Multi-rotor Unmanned Aerial Vehicles (UAVs), although originally designed and developed for defence and military purposes, in the last ten years have gained momentum, especially for civilian applications, such as search and rescue, surveying and mapping, and agricultural crops and monitoring. Thanks to their hovering and Vertical Take-Off and Landing (VTOL) capabilities and the capacity to carry out tasks with complete autonomy, they are now a standard platform for both research and industrial uses. However, while the flight control architecture is well established in the literature, there are still many challenges in designing autonomous guidance and navigation systems to make the UAV able to work in constrained and cluttered environments or also indoors. Therefore, the main motivation of this work is to provide a comprehensive and exhaustive literature review on the numerous methods and approaches to address path-planning problems for multi-rotor UAVs. In particular, the inclusion of a review of the related research in the context of Precision Agriculture (PA) provides a unified and accessible presentation for researchers who are initiating their endeavours in this subject.

The presence of a soil seedbank facilitates the persistence of annual weed species in arable fields. Soil weed seedbank is replenished by many sources, but the largest one is the seeds produced by uncontrolled late-season weed escapes. The estimation of weed seed production potential from late-season escapes may allow farmers to make appropriate management decisions to minimize seedbank replenishment. The objective of this research was to evaluate the feasibility of using unmanned aerial vehicle–based RGB and multispectral imagery for estimating seed rain potential in late-season weed escapes in crop fields. Three case studies were used to capture images of weed escapes before crop harvest: common waterhemp [Amaranthus tuberculatus (Moq.) Sauer] in soybean [Glycine max (L.) Merr.], Palmer amaranth [Amaranthus palmeri (S.) Watson] in cotton (Gossypium hirsutum L.), and johnsongrass [Sorghum halepense (L.) Pers.] in soybean. Randomly selected quadrats with different density gradients of weed escapes were sampled at the time of crop maturity. High-resolution RGB and multispectral images of the experimental area were collected using drones immediately before ground sample collection. Normalized difference vegetation index (NDVI), excess green index (ExG), and canopy volume estimates derived from canopy height models were used to obtain weed biological measurements (biomass and seed production). Among the indices investigated, NDVI and ExG had very strong correlations (0.71 to 0.97) with weed biomass. No specific remote sensing variable was ideal across the three cases examined here, suggesting that a generalized remote sensing approach may not offer robust estimations and case-specific applications are imperative. Nonetheless, drone imagery is a powerful tool for estimating seed production from uncontrolled weed escapes and assisting with management decision making.

A thorough understanding of commonly used herbicide application practices and technologies is needed to provide recommendations and determine necessary application education efforts. An online survey to assess ground and aerial herbicide application practices in Arkansas was made available online in spring 2019. The survey was direct-emailed to 272 agricultural aviators and 831 certified commercial pesticide applicators, as well as made publicly available online through multiple media sources. A total of 124 responses were received, of which 75 responses were specific to herbicide applications in Arkansas agronomic crops, accounting for approximately 49% of Arkansas’ planted agronomic crop hectares in 2019. Ground and aerial application equipment were used for 49% and 51% of the herbicide applications on reported hectares, respectively. Rate controllers were commonly used application technologies for both ground and aerial application equipment. In contrast, global positioning system-driven automatic nozzle and boom shut-offs were much more common on ground spray equipment than aerial equipment. Applicator knowledge of nozzles and usage was limited, regardless of ground or aerial applicators, as only 28% of respondents provided a specific nozzle type used, indicating a need for educational efforts on nozzles and their importance in herbicide applications. Of the reported nozzle types, venturi nozzles and straight-stream nozzles were the most commonly used for ground and aerial spray equipment, respectively. Spray carrier volumes of 96.3 and 118.8 L ha−1 for ground spray equipment and 49.6 and 59.9 L ha−1 for aerial application equipment were the means of reported spray volumes for systemic and contact herbicides, respectively. Respondents indicated application optimization was a major benefit of utilizing newer application technologies, herbicide drift was a primary challenge, and research needs expressed by respondents included adjuvants, spray volume efficacy, and herbicide drift. Findings from this survey provided insight into current practices, technologies, and needs of Arkansas herbicide applicators. Research and education efforts can be implemented as a result to address aforementioned needs while providing applied research-based information to applicators based on current practices.

In recent years, unmanned aerial vehicle (UAV) technology has expanded to include UAV sprayers capable of applying pesticides. Very little research has been conducted to optimize application parameters and measure the potential of off-target movement from UAV-based pesticide applications. Field experiments were conducted in Raleigh, NC during spring 2018 to characterize the effect of different application speeds and nozzle types on target area coverage and uniformity of UAV applications. The highest coverage was achieved with an application speed of 1 m s−1 and ranged from 30% to 60%, whereas applications at 7 m s−1 yielded 13% to 22% coverage. Coverage consistently decreased as application speed increased across all nozzles, with extended-range flat-spray nozzles declining at a faster rate than air-induction nozzles, likely due to higher drift. Experiments measuring the drift potential of UAV-applied pesticides using extended-range flat spray, air-induction flat-spray, turbo air–induction flat-spray, and hollow-cone nozzles under 0, 2, 4, 7, and 9 m s−1 perpendicular wind conditions in the immediate 1.75 m above the target were conducted in the absence of natural wind. Off-target movement was observed under all perpendicular wind conditions with all nozzles tested but was nondetectable beyond 5 m away from the target. Coverage from all nozzles exhibited a concave-shaped curve in response to the increasing perpendicular wind speed due to turbulence. The maximum target coverage in drift studies was observed when the perpendicular wind was 0 and 8.94 m s−1, but higher turbulence at the two highest perpendicular wind speeds (6.71 and 8.94 m s−1) increased coverage variability, whereas the lowest variability was observed at 2.24 m s−1 wind speed. Results suggested that air-induction flat-spray and turbo air–induction flat-spray nozzles and an application speed of 3 m s−1 provided an adequate coverage of target areas while minimizing off-target movement risk.

Chemical weed control remains a widely used component of integrated weed management strategies because of its cost-effectiveness and rapid removal of crop pests. Additionally, dicamba-plus-glyphosate mixtures are a commonly recommended herbicide combination to combat herbicide resistance, specifically in recently commercially released dicamba-tolerant soybean and cotton. However, increased spray drift concerns and antagonistic interactions require that the application process be optimized to maximize biological efficacy while minimizing environmental contamination potential. Field research was conducted in 2016, 2017, and 2018 across three locations (Mississippi, Nebraska, and North Dakota) for a total of six site-years. The objectives were to characterize the efficacy of a range of droplet sizes [150 µm (Fine) to 900 µm (Ultra Coarse)] using a dicamba-plus-glyphosate mixture and to create novel weed management recommendations utilizing pulse-width modulation (PWM) sprayer technology. Results across pooled site-years indicated that a droplet size of 395 µm (Coarse) maximized weed mortality from a dicamba-plus-glyphosate mixture at 94 L ha–1. However, droplet size could be increased to 620 µm (Extremely Coarse) to maintain 90% of the maximum weed mortality while further mitigating particle drift potential. Although generalized droplet size recommendations could be created across site-years, optimum droplet sizes within each site-year varied considerably and may be dependent on weed species, geographic location, weather conditions, and herbicide resistance(s) present in the field. The precise, site-specific application of a dicamba-plus-glyphosate mixture using the results of this research will allow applicators to more effectively utilize PWM sprayers, reduce particle drift potential, maintain biological efficacy, and reduce the selection pressure for the evolution of herbicide-resistant weeds.

Preemergence herbicides are typically applied by broadcasting to the top of raised beds before laying the plastic mulch in plasticulture production systems. Broadleaf and grass emergence is limited to transplant holes in the mulch. As a result, most herbicides are applied under the mulch in locations where weeds cannot emerge and herbicides are unnecessary. To reduce this excessive off-target application, a precision hole-punch sprayer was developed at the University of Florida for use in plasticulture production systems. The technology facilitates the application of herbicides during the hole-punch operation immediately before transplant. Application of napropamide and S-metolachlor in an application volume of 233 L ha−1 of water using the precision hole-punch applicator had no effect on tomato and bell pepper growth and yield. Equipment accuracy ranged from 55% to 90%. Preemergence herbicide use was reduced by 88% to 92% with no reduction in weed control. The hole-punch applicator is an effective way to reduce PRE herbicide use in transplant vegetables grown using the plasticulture production system.

Nearly two decades into the 21st century, we revisit the topic of changes in the US agricultural system. We focus on trends in structure, technology and policy, and on the increasing influence of consumer preferences on this system, particularly for organic agriculture and local and regional foods. We examine technological innovations in the 21st century, including biotechnology, precision agriculture and indoor farming. Within overall trends toward consolidation, we identify an increasing number of vegetable farms and greenhouse operations, accompanied by a decrease in average size of those operations. We note the shift away from price support toward greater reliance on risk management in farm policy, and also track the impact of food movement trends on recent farm bills. While farm bill policies continue to focus on conventional field crop agriculture, some trends—expanded crop insurance, conservation program support and spending on federal data collection, research and community-based grants, for example—have begun to incorporate the growing movement toward organic, local and regional food systems into the mainstream of US agricultural production and policy.

The objective of this research was to evaluate producers’ perspectives of four key precision agriculture technologies (variable rate fertilizer application, precision soil sampling, guidance and autosteer, and yield monitoring) in terms of the benefits they provide to their farms (increased yield, reduced production costs, and increased convenience) using a best-worst scaling choice experiment. Results indicate that farmers’ perceptions of the benefits derived from various precision agriculture technologies are heterogeneous. To better understand farmers’ adoption decisions, or lack thereof, it is important to first understand their perceptions of the benefits precision agriculture technologies provide.

The discipline of weed science is at a critical juncture. Decades of efficient chemical weed control have led to a rise in the number of herbicide-resistant weed populations, with few new herbicides with unique modes of action to counter this trend and often no economical alternatives to herbicides in large-acreage crops. At the same time, the world population is swelling, necessitating increased food production to feed an anticipated 9 billion people by the year 2050. Here, we consider these challenges along with emerging trends in technology and innovation that offer hope of providing sustainable weed management into the future. The emergence of natural product leads in discovery of new herbicides and biopesticides suggests that new modes of action can be discovered, while genetic engineering provides additional options for manipulating herbicide selectivity and creating entirely novel approaches to weed management. Advances in understanding plant pathogen interactions will contribute to developing new biological control agents, and insights into plant–plant interactions suggest that crops can be improved by manipulating their response to competition. Revolutions in computing power and automation have led to a nascent industry built on using machine vision and global positioning system information to distinguish weeds from crops and deliver precision weed control. These technologies open multiple possibilities for efficient weed management, whether through chemical or mechanical mechanisms. Information is also needed by growers to make good decisions, and will be delivered with unprecedented efficiency and specificity, potentially revolutionizing aspects of extension work. We consider that meeting the weed management needs of agriculture by 2050 and beyond is a challenge that requires commitment by funding agencies, researchers, and students to translate new technologies into durable weed management solutions. Integrating old and new weed management technologies into more diverse weed management systems based on a better understanding of weed biology and ecology can provide integrated weed management and resistance management strategies that will be more sustainable than the technologies that are now failing.

Precision weed management, an application of precision agriculture, accounts for within-field variability of weed infestation and herbicide damage. Unmanned aerial vehicles (UAVs) provide a unique platform for remote sensing of field crops. They are more efficient and flexible than manned agricultural airplanes in acquiring high-resolution images at low altitudes and low speeds. UAVs are more universal than agricultural aircraft, because the latter are used only in specific regions. We have developed and used UAV systems for red–green–blue digital and color–infrared imaging over crop fields to identify weed species, determine crop injury from dicamba at different doses, and detect naturally grown glyphosate-resistant weeds. This article presents remote sensing technologies for weed management and focuses on development and application of UAV-based low-altitude remote sensing technology for precision weed management. In particular, this article futher discusses the potential application of UAV-based plant-sensing systems for mapping the distributions of glyphosate-resistant and glyphosate-susceptible weeds in crop fields.

Three methods of predicting the impact of weed interference on crop yield and expected economic return were compared to evaluate the economic importance of weed spatial heterogeneity. Density of three weed species was obtained using a grid sampling scheme in 11 corn and 11 soybean fields. Crop yield loss was predicted assuming densities were homogeneous, aggregated following a negative binomial with known population mean and k, or aggregated with weed densities spatially mapped. Predicted crop loss was lowest and expected returns highest when spatial location of weed density was utilized to decide whether control was justified. Location-specific weed management resulted in economic gain as well as a reduction in the quantity of herbicide applied.