Agricultural Tasks Research Articles

Towards an Architecture for Online Scheduling of Autonomous Robots in Agriculture

Nowadays, we observe the development of autonomous robots for agricultural tasks. Farmers are becoming task and data managers with the emergence of advanced farm management information systems (FMISs). However, existing FMISs lack the tools for handling scheduling and monitoring of fleets of robots. The scheduling functionalities are essential for the growth of autonomous robot industry. It allows a better management to share these state of the art and expensive resources between multiple farmers, reducing the overall cost. Scheduling is always coupled with a re-scheduling process that allows to react to unexpected events. The re-scheduling process, called online scheduling, can only be made possible with a monitoring process that collects real-time information about the ongoing tasks and the state of robots. Finally, relatively little is known about the changes in farmers' activities as a result of the introduction of these robots. Acceptance of these new technologies is nevertheless essential to the performance of the systems. Motivated by the lack of a general framework for the online scheduling of autonomous robots for agriculture, the authors propose a conceptual framework for the scheduling and monitoring of such shared resources. All the needed building blocks for the whole conceptual framework to function efficiently are highlighted. Open issues related to each of these building blocks are discussed, from robotics auto-diagnosis to data management, wireless communication, scheduling, monitoring, and controlling these autonomous robots, keeping in the loop the human operator and his essential role in this system.

Crop Yield Prediction in Precision Agriculture

Predicting crop yields is one of the most challenging tasks in agriculture. It plays an essential role in decision making at global, regional, and field levels. Soil, meteorological, environmental, and crop parameters are used to predict crop yield. A wide variety of decision support models are used to extract significant crop features for prediction. In precision agriculture, monitoring (sensing technologies), management information systems, variable rate technologies, and responses to inter- and intravariability in cropping systems are all important. The benefits of precision agriculture involve increasing crop yield and crop quality, while reducing the environmental impact. Simulations of crop yield help to understand the cumulative effects of water and nutrient deficiencies, pests, diseases, and other field conditions during the growing season. Farm and in situ observations (Internet of Things databases from sensors) together with existing databases provide the opportunity to both predict yields using “simpler” statistical methods or decision support systems that are already used as an extension, and also enable the potential use of artificial intelligence. In contrast, big data databases created using precision management tools and data collection capabilities are able to handle many parameters indefinitely in time and space, i.e., they can be used for the analysis of meteorology, technology, and soils, including characterizing different plant species.

Algorithms for Calculating the Trajectory of Unmanned Aerial Vehicles for Solving Agricultural Problems

The relevance of using unmanned aerial vehicles (UAV) is substantiated in comparison with other methods of aerospace survey. The paper provides examples of tasks requiring the use of UAVs for aerial photography from different heights. It is shown that the introduction of agricultural robots, including UAVs, increases the speed of fi eld work, allows obtaining unique data necessary for the assessment of agricultural territories, crops processing and plant growth stimulation. It is noted that the problem of constructing the movement trajectories of a multirotor UAV for performing agricultural tasks within a minimum time limit remains unresolved. (Research purpose) To reduce the length of the trajectory covering a given area and reduce the flight time of a multirotor UAV, taking into account the analysis of possible obstacles and land plots that are beyond the task scope. (Materials and methods) Geometric methods have been used to calculate the UAV flight trajectory covering a given section, the trajectory of movement in an environment with obstacles to the designated point. Photogrammetry methods have been used for processing aerial photography images when forming an orthophotoplane and a terrain map. (Results and discussion) The trajectory calculated by the developed algorithm proves to meet all the requirements: it is continuous, has a minimum number of turns, it is smoothed, and feasible for a multirotor UAV. (Conclusions) It was determined that according to the proposed algorithm it takes less than 0.05 seconds to calculate the trajectory covering a rectangular section with the sides of 200 by 30 meters. It was found that the trajectory in the fi rst 10,200-square-meter section decreased by 9 percent, and in the second 950,000-square-meter section it reduced by 6 percent, compared with the length of the trajectory built using standard algorithms. The flight time reduced by 32 and 10 percent, respectively. The paper presents the key advantages of using UAV for video shooting such as: guaranteed high resolution of photographic materials and the ability to shoot at a given time, allowing for the crop condition assessment.

Regional Time Series Forecasting of Chickpea using ARIMA and Neural Network Models in Central Plains of Uttar Pradesh (India)

Climate and yield prediction are the most important and challenging tasks in modern agriculture during the climate change era. In general, climate and yield are mostly non-linear and highly complicated. India is an agricultural country and most of its economy depends upon agriculture therefore early prediction of climate and yield is necessary for the planned economic growth of our country. This research identifies superior forecasting models of Autoregressive Integrated Moving Average (ARIMA) as well as Artificial Neural Network (ANN) for predicting future climate and chickpea yield. Historical data for the climate and crop were used (1996-2020) and forecasting was done for the next 5 years (2020-2025). By using, RMSE and R2 statistical tools simultaneously, the predictive accuracy of ARIMA and ANN models was compared. By comparing the R2 values of ARIMA (0.591) and ANN (0.96), this study reveals ANN models can be used as more accurate forecasting tools to predict the future climate as well as yield, enabling timely agricultural management.

Bee Sound Detector: An Easy-to-Install, Low-Power, Low-Cost Beehive Conditions Monitoring System

One of the most significant agricultural tasks in beekeeping involves continually observing the conditions inside and outside the beehive. This is mainly performed for the early detection of some harmful events. There have been many studies on how to detect and prevent such occurrences by performing periodic interventions or, when the frequency of such actions is hard to enforce, by using sensory systems that record the temperature, humidity, and weight of the beehive. Nevertheless, such methods are inaccurate, and their delivered outcomes usually diverge from the actual event or false trigger and introduce more effort and damage. In this paper, the authors propose a new low-cost, low-power system called Bee Sound Detector (BeeSD). BeeSD is a low-cost, embedded solution for beehive quality control. It incorporates the sensors mentioned above as well as real-time sound monitoring. With the combination of temperature, humidity, and sound sensors, the BeeSD can spot Colony Collapse Disorder events due to famine and extreme weather events, queen loss, and swarming. Furthermore, as a system, the BeeSD uses cloud logging and an appropriate mobile phone application to push notifications of extreme measurements to the farmers. Based on achieved performance indicators, the authors present their BeeSD IoT device and system operation, focusing on its advantages of low-cost, low-power, and easy-to-install characteristics.

Conceptualization and Implementation of a Reconfigurable Unmanned Ground Vehicle for Emulated Agricultural Tasks

Small-to-medium sized systems able to perform multiple operations are a promising option for use in agricultural robotics. With this in mind, we present the conceptualization and implementation of a versatile and modular unmanned ground vehicle prototype, which is designed on top of a commercial wheeled mobile platform, in order to test and assess new devices, and motion planning and control algorithms for different Precision Agriculture applications. Considering monitoring, harvesting and spraying as target applications, the developed system utilizes different hardware modules, which are added on top of a mobile platform. Software modularity is realized using the Robot Operating System (ROS). Self- and ambient-awareness, including obstacle detection, are implemented at different levels. A novel extended Boundary Node Method is used for path planning and a modified Lookahead-based Line of Sight guidance algorithm is used for path following. A first experimental assessment of the system’s capabilities in an emulated orchard scenario is presented here. The results demonstrate good path-planning and path-following capabilities, including cases in which unknown obstacles are present.

Robust Backstepping Control Applied to UAVs for Pest Recognition in Maize Crops

In this paper, a robust control technique is developed to achieve the quadrotor stabilization against unmodeled matching vanishing dynamics. The synthesis of the proposed robust control is based on the Lyapunov approach and the backstepping method allowing to construct an iterative control algorithm. To compare the performance of the proposed controller, a Proportional Derivative (PD) controller is used to obtain experimental results in an outdoor environment. To compare the closed-loop system responses with both controllers, the Integral Absolute Error is computed and several tests are conducted to calculate the error standard deviation. Ultimately, employing the robust backstepping control approach in pest recognition in maize crops, which is a specific task of precision agriculture, demonstrates its effectiveness in improving the trajectory tracking of the vehicle while it captures images of the crops.

Classification of Weeds of Paddy Fields using Deep Learning

Weed management is one of the important tasks in agriculture. Weeds in rice fields are usually managed using three ways - chemical herbicides, mechanical weeders, and manual weeding. Manual weeding becomes a problem when there is a shortage of agricultural laborers. Mechanical weeders are not suitable for direct-seeded rice fields. Chemical herbicides are not advisable especially when farmers do not know about site-specific weed management. Site-specific weed management is using the right herbicide in the right amount. Therefore, this paper investigates computer vision-based deep learning techniques with transfer learning classifying three types of weeds in paddy fields, namely sedges, grasses, and broadleaved weeds so that the right herbicide is recommended to the farmers. This would reduce the broadcast application and the overuse of the herbicides, thereby limiting the negative impact of the chemical herbicides on the environment. This research work shows promising results with an accuracy around 90% and thus encourages the development of digital agriculture.

CEIFA: A multi-level anomaly detector for smart farming

Climate change, the water crisis, and population growth add new challenges for food production. The modernization of agricultural methods is essential to increase production rates and preserve natural resources. Smart agriculture provides resources that can enhance farming tasks by efficiently controlling actuators, optimizing utility and resource use, managing production, maximizing profit, and minimizing costs. For these technologies to become popular, they must have a high level of reliability and safety. To improve the reliability in Smart Agriculture, this paper proposes CEIFA, a low-cost, hybrid anomaly detector capable of identifying failures, faults, errors and attacks that impact these systems. CEIFA operates on local or remote cloud servers, filtering data sent by agricultural system sensors. It can operate on resource-restricted devices and save financial resources related to computing costs. Real tests on a set of faults, failures, and errors point to an efficiency greater than 95% in anomaly detection.

Dancers, Mountains, and People in the Fields

Dancers, Mountains, and People in the Fields

Design and Development of IoT Based Remote Sensing System for Smart Farming

Abstract: Now a day’s a remote sensing system in IoT based environment widely used in smart farming. This system in smart farming is used to perform various farming task like soil sensing, temperature sensing, weather sensing, light sensing, and moisture sensing smarter and efficient for better yielding and production. The proposed system aims to design and develop a remote sensing tool that senses the current status of plant and disease in the plants in farm field using various sensors like temperature, humidity, pH sensors and water level sensors. Identification of plant disease is done through sensor readings data. The main aim of the research paper is to design and development of an Internet of Things (IoT) based remote sensing system tool to supervise and control the citrus plant in farm field. This proposed system consists of hardware setup with real-time sensors such as soil moisture, sunlight, humidity, temperature, water level, and camera module which are controlled by Raspberry Pi processor with Python and configured essential getaways to sense and stored data associated to parameters for supervising the dynamic plant growth. With the help of such kind of IoT based remote sensing system farmer can observe all the parameters of plant remotely and get real time analysis of data, so that they can increase the productivity.

A Multi-Control Strategy to Achieve Autonomous Field Operation

Nowadays, there are several methods of controlling a robot depending on the type of agricultural environment in which it operates. In order to perform a complete agricultural task, this paper proposes a switching strategy between several perception/control approaches, allowing us to select the most appropriate one at any given time. This strategy is presented using an electrical tractor and three control approaches we have developed: path tracking, edge following and furrow pursuing. The effectiveness of the proposed development is tested through full-scale experiments in realistic field environments, performing autonomous navigation and weeding operations in an orchard and an open field. The commutation strategy allows us to select behavior depending on the context, with a good robustness with respect to different sizes of crops (maize and bean). The accuracy stays within ten centimeters, allowing us to expect the use of robots to help with the development of agroecological principles.

Village level identification of sugarcane in Sangali, Maharashtra using open source data

Agriculture and crop monitoring are very important for an agrarian country like India. This study is done in June Khed village in the Sangli district of Maharashtra, India to assessing the efficiency of an open source cloud-based remote sensing platform Google Earth Engine (GEE), in the village-scale identification of sugarcane. The ground-truth data was collected and the efficiency of Landsat-8 and Sentinel-2 satellite data was assessed in driving GEE’s inbuilt Machine Learning classifiers: Classification and Regression Tree (CART), Support Vector Machine (SVM) and Random Forest (RF). Results were validated with the ground truth data and the official data. Of the methods used, SVM outperformed RF and CART with the lowest relative deviation (+9.2%), highest F1- score (0.8) and overall accuracy (78%), using the Sentinel-2 data. Results also indicated the in-situ use of observation data with high spatio-temporal resolution data. The validated model was then up-scaled for the Walwa Taluka level, to map sugarcane area that can be used for further agriculture tasks such as crop monitoring and yield prediction, leading to better management of crop and better formulating of sugarcane farmer policy.

An Overview of End Effectors in Agricultural Robotic Harvesting Systems

In recent years, the agricultural sector has turned to robotic automation to deal with the growing demand for food. Harvesting fruits and vegetables is the most labor-intensive and time-consuming among the main agricultural tasks. However, seasonal labor shortage of experienced workers results in low efficiency of harvesting, food losses, and quality deterioration. Therefore, research efforts focus on the automation of manual harvesting operations. Robotic manipulation of delicate products in unstructured environments is challenging. The development of suitable end effectors that meet manipulation requirements is necessary. To that end, this work reviews the state-of-the-art robotic end effectors for harvesting applications. Detachment methods, types of end effectors, and additional sensors are discussed. Performance measures are included to evaluate technologies and determine optimal end effectors for specific crops. Challenges and potential future trends of end effectors in agricultural robotic systems are reported. Research has shown that contact-grasping grippers for fruit holding are the most common type of end effectors. Furthermore, most research is concerned with tomato, apple, and sweet pepper harvesting applications. This work can be used as a guide for up-to-date technology for the selection of suitable end effectors for harvesting robots.

AN OPEN-SOURCE MOBILE GEOSPATIAL PLATFORM FOR AGRICULTURAL LANDSCAPE MAPPING: A CASE STUDY OF WALL-TO-WALL FARM SYSTEMS MAPPING IN TONGA

Abstract. Pacific Island Countries (PICs) such as Tonga rely on services provided by agricultural landscapes to support livelihoods, economic activity, and food security. At the same time these landscapes face numerous pressures and risks from factors such as environmental, climate, and market changes. Accurate, spatially explicit, and timely datasets on agricultural systems is required for an array of land and agricultural management tasks. Here, the development of an open-source ICT system providing geospatial tools for landscape monitoring, developed in collaboration between geospatial researchers and Tonga’s Ministry of Agriculture, Food, and Forests (MAFF), is presented. The agile and iterative ICT for Development (ICT4D) framework used to elicit MAFF’s requirements for the ICT system is presented alongside the system architecture and case studies demonstrating its impact. A key goal of the ICT4D development process was to develop an ICT system to support MAFF from transitioning from infrequent, paper-based farm surveys to coordinated, large-team, spatially explicit digital surveying augmented by tools for analysis and reporting. The mature system architecture which includes QField and QFieldCloud, and new open-source geospatial components for spatial visualisation, analysis, and reporting is presented. Case studies where the mature tool was used by MAFF’s are presented and include: (1) how a large survey team captured spatial data for >11,000 farms for country-wide farm monitoring; and (2) how the tool informed MAFF’s landscape decision making including recovery efforts after the 2022 Hunga Tonga–Hunga Ha’apai volcano explosion. The success of the tool demonstrates the importance of stakeholder engagement and the great potential for open-source geospatial tools for landscape management and disaster response in PICs.

In-Field Automatic Identification of Pomegranates Using a Farmer Robot

Ground vehicles equipped with vision-based perception systems can provide a rich source of information for precision agriculture tasks in orchards, including fruit detection and counting, phenotyping, plant growth and health monitoring. This paper presents a semi-supervised deep learning framework for automatic pomegranate detection using a farmer robot equipped with a consumer-grade camera. In contrast to standard deep-learning methods that require time-consuming and labor-intensive image labeling, the proposed system relies on a novel multi-stage transfer learning approach, whereby a pre-trained network is fine-tuned for the target task using images of fruits in controlled conditions, and then it is progressively extended to more complex scenarios towards accurate and efficient segmentation of field images. Results of experimental tests, performed in a commercial pomegranate orchard in southern Italy, are presented using the DeepLabv3+ (Resnet18) architecture, and they are compared with those that were obtained based on conventional manual image annotation. The proposed framework allows for accurate segmentation results, achieving an F1-score of 86.42% and IoU of 97.94%, while relieving the burden of manual labeling.

Many-objective evolutionary algorithm based agricultural mobile robot route planning

Agricultural robot technology has experienced rapid development in the past ten years, and agricultural robots have been used to implement various complex agricultural tasks. In these processes, route planning is an important guarantee for reducing navigation distance and saving total turning angle. However, minimizing the cost of the entire navigation process on the premise of completing agricultural work is difficult. Many-objective Evolutionary Algorithm is used to solve the route planning problem of agricultural mobile robots under the premise of minimizing navigation cost. By scanning the radar map of the greenhouse, the path between all target points is calculated by using the probabilistic roadmap (PRM), and the route planning of the agricultural robot is carried out according to the sum of the path length and the path angle. To determine the best route for agricultural mobile robots, four algorithms are compared: Hypervolume Estimation Algorithm (HypE), Grid-Based Evolutionary Algorithm (GrEA), Knee Point-Driven Evolutionary Algorithm (KnEA), and Non-dominated sorting genetic algorithm (NSGA-III). The quality of the solutions was compared using C-Metric, and it could verify that HypE offers the best performance among four algorithms.

Prospects for the use of artificial intelligence in the agricultural sector

The relevance of the study is due to the fact that there is a rapid development of artificial intelligence technologies, due to breakthroughs in the digital environment and the uncertainty of their application in the agricultural sector. It should be noted that Russia plays an important role in the world history for the development of these technologies, but the Soviet management model limits the rapid development of digital technologies in modern Russia. Artificial intelligence technologies are of particular importance, as they are used for big data analysis, robotics and as the basis for various forecasting algorithms. The article reveals the role of artificial intelligence in solving complex tasks in agriculture, identifies the main barriers to the introduction of digital technologies and identifies perspective direction of application for artificial intelligence technologies. Due to the lack of scientific developments on this issue, the lack of research on the feasibility and effectiveness of the using artificial intelligence, the lack of methodological recommendations for the introduction of these technologies and systematic training of artificial intelligence, it is difficult to use artificial intelligence in agricultural organizations. For Russia, the issue of agricultural automation is particularly important: over the past 10 years, the number of workers in the industry has more than halved. In the future, automated systems can improve rural areas and keep production there, since most of the rural population is no longer engaged in agriculture. Agriculture remains one of the most important and largest industries, and even a small increase in efficiency will lead to a significant increase due to its scale.

Closed Spontaneous Rupture of the Flexor Tendon in Zone 3: A Case Series

Purpose: The pathophysiology of spontaneous rupture is not fully understood. Spontaneous rupture has been questioned, as some authors have stated that most spontaneous ruptures cannot be attributed to a cause or have not yet been investigated. Zone 3 was the most common place for flexor tendon ruptures to happen on their own, but the risk factor was not found. Methods: From 2019 to 2022, we recorded four cases of closed rupture of the flexor tendon in zone 3 caused by agricultural activity. The history of all patients such as steroid injections, pain, and underlying conditions, was reviewed. Some parts of the tendon, debrided from the injury site, were sent for pathologic examination. The final range of motion and activity were recorded. Results: All our patients had acute symptoms while gripping tools and excavating dirt during gardening. All flexor tendon ruptures occurred in zone 3, and were repaired by core and epitendinous suture. There was no actual condition believed to be the cause of the rupture. All patients achieved nearly full range of motion in the final follow-up. Conclusions: A firm grip while performing agricultural tasks appears to be a risk factor of closed rupture flexor tendon. Although it is uncommon for a closed rupture of a flexor tendon injury to occur without ascertainable causes, this condition should be recognized while facing weakness or an inability to flex the finger.

A Review of Current and Historical Research Contributions to the Development of Ground Autonomous Vehicles for Agriculture

In this study, a comprehensive overview of the available autonomous ground platforms developed by universities and research groups that were specifically designed to handle agricultural tasks was performed. As cost reduction and safety improvements are two of the most critical aspects for farmers, the development of autonomous vehicles can be of major interest, especially for those applications that are lacking in terms of mechanization improvements. This review aimed to provide a literature evaluation of present and historical research contributions toward designing and prototyping agricultural ground unmanned vehicles. The review was motivated by the intent to disseminate to the scientific community the main features of the autonomous tractor named BOPS-1960, which was conceived in the 1960s at the Alma Mater Studiorum University of Bologna (UNIBO). Jointly, the main characteristics of the modern DEDALO unmanned ground vehicle (UGV) for orchard and vineyard operations that was designed recently were evaluated. The basic principles, technology and sensors used in the two UNIBO prototypes are described in detail, together with an analysis of UGVs for agriculture conceived in recent years by research centers all around the world.