Consistent lidar-only SLAM for legged agricultural robots in arboreal environments via robust dimensionality reduction

Modern robotic systems address complex engineering challenges using artificial intelligence and machine learning techniques. In agricultural robotics, fruit identification and sorting remain challenging due to variations in size, shape, color, orientation, and lighting conditions. This study presents the design and implementation of a vision-based fruit sorting robotic system integrating YOLOv8-based object detection with robotic manipulation. A custom dataset consisting of images of 2 different fruits (namely banana and strawberry images), including single-fruit and multi-fruit scenarios, was used and manually annotated using bounding boxes in CVAT. The dataset was divided into training, validation, and test subsets to enable robust model development under realistic operational conditions. A lightweight YOLOv8 model was trained using CUDA acceleration and optimized for edge deployment by selecting YOLOv8n to balance inference speed and detection accuracy. The trained model was converted to ONNX format and deployed on a Raspberry Pi 5 for real-time inference using live camera input. Evaluation on an independent test dataset achieved a precision of 0.999, recall of 1.000, mAP@0.5 of 0.995, and mAP@0.5:0.95 of 0.963 under controlled experimental conditions with limited object classes. The modular architecture enables low-cost and scalable deployment and provides a foundation for future enhancements, including closed-loop robotic control, additional object categories, and operation in more dynamic environments.

The automated assessment of tomato ripeness is vital for modern greenhouse operations, yet challenges remain due to variable environmental conditions. To provide a solution, we propose rank-aware You Only Look Once (YOLO), a novel detection framework that incorporates the biological prior of top-to-bottom ripening within fruit clusters. This is achieved through two key innovations: an efficient position-aware head for regressing relative height for fruits and a dynamic margin-aware ranking loss (DM-RankLoss) that enforces the correct spatial sequence. Evaluated on a 3500-image dataset from a solar greenhouse, our plug-and-play module could boost the mean average precision (mAP) at intersection over union (IoU) threshold of 0.50 (mAP50) of multiple YOLO architectures by up to 5.66 pecentage points. The model effectively learns the cluster topology, achieving a height-mean absolute error (H-MAE) of 0.107 (normalized) and a pairwise ranking accuracy (PRA) of 84.59%, while it reduces the parameter count by over 10% compared to the baseline for efficient deployment. Visualizations confirm that the model leverages spatial context to resolve color ambiguities. Our work offers a sensor-free, accurate, and efficient solution for in situ phenotyping in agricultural robotics.

Modern agriculture faces a convergence of critical challenges: acute labour shortages, escalating agrochemical costs, inefficient manual seed sowing, and persistent weed infestations that collectively reduce crop yields by 20 - 40% in smallholder farms. This paper presents a Multipurpose Agriculture Robot, a low-cost, farmer-configurable, IoT-controlled robotic platform designed to address these challenges through a unified modular architecture. The proposed system integrates three primary operational units - chemical spraying and irrigation unit, a precision seeding unit and a cutting unit - mounted on a common ESP32-based chassis equipped with servo-actuated extensible folding-arm mechanisms. The arms dynamically adjust irrigation and pesticide spray coverage width from 30 cm to 90 cm per side in real time without halting field operations, a feature not available in any existing low-cost agricultural robot. Optional attachments including a field - leveling tool can be added or removed via a standardized quick-connect modular tool bay, enabling season-specific farmer configuration. A dedicated mobile application communicates with the robot over Bluetooth and Wi-Fi, providing real-time directional control, arm angle adjustment, spray activation and cutting unit. Mathematical models govern irrigation water calculation using soil moisture feedback and seeding error minimization using motor-speed adjustment.

ABSTRACT The selective migration of young people to urban centres has resulted in an ageing workforce within the Brazilian coffee sector, creating a growing demand for solutions that aid producers in crop management routines. Coffee production, the fourth largest in terms of gross national agricultural revenue, can benefit from technological support. In this context, this research investigates social perceptions of a prototype of social robots by agricultural producers. The study involved a sample of 28 rural producers and stakeholders, stratified by age into two independent groups: 18–49 years () and 50+ years (). Participants evaluated the robot using the Robotic Social Attributes Scale (RoSAS), enabling a detailed look at social perceptions. The analysis yielded four distinct user profiles: (1) Engaged Supporters, (2) Pragmatic Users, (3) Innovation Sceptics and (4) Confident Adopters. Furthermore, correlation analysis highlighted strong links between perceived social attributes (e.g., high correlation between trustworthy and competent). Crucially, non‐parametric testing revealed a statistically significant difference across age groups regarding the robot's perceived organic nature (). These findings indicate a clear generational shift in social robot acceptance, underscoring the need for age‐sensitive design in future assistive robotics for agriculture.

Accurate tomato ripeness detection is crucial for automated harvesting; however, complex greenhouse environments—characterized by dynamic light interference, foliage occlusion, and dense fruit overlapping—severely hinder detection performance and lead to frequent misdetections. This study aims to develop a high-precision, lightweight detection model that simultaneously addresses these three core challenges, thereby providing a technically deployable algorithmic foundation for resource-constrained agricultural edge devices. To this end, we propose CFD-DETR, a lightweight tomato ripeness detection model based on the RT-DETR architecture. The model incorporates a CAEfficientViT backbone for the lightweight extraction of multi-scale color and texture features. Furthermore, a Focused Efficient Additive Attention (FEAA) mechanism is integrated to capture fine-grained local ripening traits with minimal computational overhead. During feature reconstruction, a Deep Dynamic Upsampling (DwDySample) operator is utilized to preserve semantic integrity. Additionally, we designed the Wise-SIoU loss function, which dynamically penalizes low-quality samples to enhance boundary fitting and robustness against background noise. Experimental evaluations demonstrate that CFD-DETR achieves 90.2% mAP@0.5, outperforming the baseline model by 2.1 percentage points while significantly reducing the parameter count and computational complexity by 47.2% and 52.5%, respectively. Cross-dataset validation on the publicly available Laboro Tomato and RaUTD datasets confirms the model’s superior generalization capabilities. Overall, CFD-DETR provides a highly efficient and robust solution for real-time agricultural robotics.

The integration of artificial intelligence (AI) and robotics in agriculture has opened new possibilities for precision farming and sustainable crop management. This study proposes an AI-based weed detection and herbicide spraying robot designed to identify and selectively eliminate weeds with minimal human intervention. The system utilizes computer vision and machine learning algorithms to distinguish between crops and weeds in real time, enabling targeted herbicide application. By reducing excessive chemical usage, the proposed solution aims to lower environmental impact, improve crop yield, and enhance operational efficiency.The research adopts a design and experimental approach, where image datasets are used to train classification models for weed detection. The robot is equipped with sensors, a camera module, and an automated spraying mechanism that activates only when weeds are detected. Performance evaluation is based on detection accuracy, spraying precision, and reduction in herbicide consumption. Results indicate that AI-driven selective spraying significantly minimizes chemical wastage while maintaining effective weed control.The study highlights the potential of intelligent agricultural robotics as a sustainable alternative to traditional weed management practices, offering economic and environmental benefits for modern farming systems.

ABSTRACT: Agriculture in developing countries faces challenges such as labor shortage, energy inefficiency, and lack of automation. This project presents the design and implementation of the Solar Based Multifunctional Agricultural Machine using Zigbee is designed to perform multiple farming operations such as ploughing, sowing, grass cutting, and water sprinkling with reduced human effort and improved efficiency. The system uses solar energy as the main power source, making it eco-friendly and cost-effective for agricultural applications. An Arduino Uno is used as the main controller to manage all operations, while Zigbee communication enables wireless control of the machine through a laptop. DC motors are used for robot movement, servo motors perform ploughing and sowing, and relay modules control the grass cutter and water sprinkler. The LM2596 voltage regulator provides stable power supply to all components. This system helps in saving time, reducing labor, and improving agricultural productivity by integrating renewable energy and wireless communication technology, making it suitable for modern smart farming applications. Keywords: Solar energy, Zigbee wireless communication, smart farming, agricultural robot, smart farming, Arduino UNO, wireless control.

Abstract—Agriculture remains a cornerstone of developing economies, yet traditional farming practices continue to rely heavily on manual labor, incurring high operational costs and productivity losses. This paper presents the design and implementation of an IoT and Sensor-Based Smartphone-Operated Multipurpose Agricultural Robotic Vehicle (SARWR) capable of performing multiple farming operations—including grass cutting, seed sowing, soil moisture monitoring, and irrigation—within a single automated platform. The system utilizes an ESP32 microcontroller as the central processing unit, interfaced with soil moisture, temperature, humidity, and Light Dependent Resistor (LDR) sensors for real-time field monitoring. An L298N dual H-bridge motor driver controls the locomotion and task-specific DC motors, while a relay module manages the water pump. The robot is wirelessly controlled through a smartphone application via Bluetooth and Wi-Fi communication protocols. Solar panels provide a renewable energy source, making the system eco-friendly and energy-independent. The proposed system was tested under various field conditions, and results demonstrate significant reductions in manual labor, improved seed placement accuracy, and effective weed removal. The integration of IoT connectivity enables remote monitoring of field parameters through a cloud dashboard, supporting precision agriculture and sustainable farming practices. Keywords—IoT, ESP32, agricultural robot, soil moisture sensor, smartphone control, L298N motor driver, precision farming, solar energy, automation, seed sowing. Previous SOLAR BASED MULTIFUNCTIONAL AGRICULTURAL MACHINE USING ZIGBEE

The 2025 novel Mechanize My Hands for War features humanoid robots for agriculture.

Abstract Smart farming is constitutes a means of facilitating the European Green Deal and the Farm to Fork strategy. However, credible sustainability results require measurable and validatable indicators, verifiable data and automation that is reliable even under field conditions. This overview study presents developments in IoT-based sensing, artificial intelligence-based analysis and autonomous robotics, and links them to EU target areas. The peer-reviewed studies (2000–2025) were retrieved from the Scopus, Web of Science, and Google Scholar databases and supplemented with key EU legal and strategic documents. The article proposes a policy-driven digital agroecological management (PDAM) framework that establishes adaptable indicators based on past trends in EU targets (pesticides, nutrients, soil, biodiversity and climate) and then develops a system of perception-analysis-implementation based on these indicators. The most effective tools support GNSS-based machine control, variable rate application, and remote sensing, while AI-based decision support tools, autonomous weed control, and digital twin field validation are still weaker. Interoperability, data governance, cybersecurity and safety regulation emerge as critical scaling constraints for auditable smart farming systems.

Farmers today spend a lot of money on machines that help them decrease labor and increase yield of crops we introduced the automatic machine called Agricultural Robot for automatic plough and seeding. This paper strives to develop a robot capable of performing operations like automatic plough, seed dispensing, and water spraying. It also provides automatic control robot using IOT module. The main component here is the AVR At mega microcontroller that supervises the entire process. Initially the robot tills the entire field and proceeds to plough, simultaneously dispensing seeds side by side. The device controlled by IOT which continuously sends data to the microcontroller. IOT module used to control the b\robot for directions. Dc geared motors used to plough the soil and releases the seed then water sprinkler automatically sprinkle water for automatic seeding in agriculture. All components are associated to micro controller arduino. ATMEGA328 micro controller used to process input and produce output by using ARDUINO IDE for Embedded C programming. Regulated power supply gives 5v of DC voltage to perform the operation.

Abstract The goal of the research paper "Crop Defect Detection Robot with Automatic Pesticide Spraying System" is to create an intelligent agricultural robot for smart farming applications by combining image processing, artificial intelligence, and the Internet of Things. Manual labor, excessive chemical use, and direct human exposure to hazardous materials are all part of traditional pesticide spraying techniques. These traditional methods are labor-intensive, ineffective, and frequently cause environmental harm because of their uncontrolled spraying. An ESP32-CAM module serves as the central vision unit in the suggested system, which takes real-time pictures of crop leaves. It detects crop diseases such as fungal infections, brown spotting, or yellow leaves through the use of image processing techniques based on artificial intelligence.Upon detecting any defect in the crops, the robot automatically triggers its pesticide spraying technique, which is applied selectively to the infected part.Farmers can check on the health of their crops and the work being done by the robot via a mobile or web application that connects with the system using its Wi-Fi capabilities, due to the incorporation of the IoT platform. With this innovative approach, costs are reduced, less manual labor is required, less pesticide is used, and sustainable practices are encouraged. The paper demonstrates how modern automation is revolutionizing traditional agriculture through smart precision agriculture utilizing low-cost embedded devices, artificial intelligence, and IoT technologies. Keywords: Internet of Things (IoT), ESP32-CAM, Smart Agriculture, Image Processing, Artificial Intelligence, Crop Disease Detection, Automatic Pesticide Spraying, Sustainable Farming.

Over the last decade, agricultural robotics and mechatronics have advanced significantly. Notable examples include the commercialization of unmanned agricultural vehicles for cereal cultivation and the development of artificial intelligence applications in agricultural image analysis rooted in field science. However, food-security-related challenges in agricultural robotics and mechatronics cannot be effectively addressed through simple extensions or combinations of existing solutions alone. Meanwhile, the importance of food security is rapidly increasing under changing environmental, economic, and societal conditions. Nevertheless, some of these challenges have not yet been recognized or formally defined, and thus remain latent. This special issue has two primary purposes. (I) It reveals previously overlooked challenges not only within agricultural fields but also in off-field infrastructures and agricultural facilities, thereby broadening the scope of agricultural robotics and mechatronics. (II) It presents the potential of “new seeds,” including novel scientific understandings of real agricultural components and systems as well as principled integrations of existing solutions with artificial intelligence and data science methodologies. Beginning with this editorial, this special issue comprises 20 papers, including five development reports and one review. We express our sincere gratitude to all the authors and reviewers. We hope that this special issue will stimulate further research and development addressing food-security-related challenges in agricultural robotics and mechatronics.

Robotics is increasingly positioned as an enabling technology for precision agriculture, where management actions must be spatially and temporally targeted under constraints on labour, input use, safety, and environmental impact. This review synthesises studies on agricultural field robotics and organises the literature along four complementary axes: task (monitoring, weeding, spraying, and harvesting), platform (UGV, UAV, gantry/fixed-structure, greenhouse robot, and hybrid systems), autonomy-stack module (perception, localisation, planning, control, actuation, safety, and human–robot interaction), and evaluation setting (lab, greenhouse, open-field single season, and open-field multi-season/multi-site). Across these dimensions, this review analyses how platform constraints shape sensing geometry, actuation capability, localisation reliability, energy/endurance, supervision burden, and safety requirements. It further examines enabling technologies that recur across tasks, including vision and multimodal perception under occlusion and illumination variability, localisation and mapping under weak or denied GNSS, uncertainty-aware planning in deformable and partially observed environments, and compliant end-effectors for contact-rich operations. Beyond cataloguing systems, this paper emphasises evaluation practice by synthesising core task-relevant metrics, comparing laboratory and field validation settings, and proposing a reporting checklist and benchmark ladder to improve reproducibility and cross-study comparability. This review identifies recurring bottlenecks in domain shift, long-term autonomy, calibration robustness, crop-safe actuation, and safety assurance near humans, and it concludes with a staged research roadmap linking near-term evaluation reform to longer-term credible multi-site autonomy. Overall, this paper provides a structured framework for interpreting agricultural robotic systems not only by application but also by deployment context, system maturity, and evaluation credibility.

Abstract This paper covers the state-of-the-art perception and control technologies in precision spraying and mowing in permanent crops. The search was performed in six different databases, resulting in 1849 publications, from which only 94 were considered for inclusion in this review. The analysis highlighted the importance of canopy characteristics in precision spraying, focusing on parameters like height, width, leaf area, and volume, primarily using LiDAR sensors. Vision sensors also complemented LiDAR-based approaches, with diverse applications such as fruit detection and disease diagnosis. Despite valuable knowledge from studies on spray coverage assessment and real-time smartphone analysis, challenges persist, including dynamic environmental factors and the different collector materials used. Moreover, the review considers the cost of Variable Rate Technology (VRT) solutions in agriculture, enhancing their impact on accessibility, adoption, and sustainability. While conventional herbicide-based weed management prevails, interest in alternative techniques like mechanical mowing and organic mulches is growing, promising improved soil health and reduced environmental impact, particularly in permanent crops. To address these challenges, agricultural robotics play a crucial role in automating precision spraying and mowing, optimizing resource usage, and increasing operational precision. This systematic review highlights the state of precision agriculture in permanent crops and emphasizes the need for continued research and development to improve the sustainability and efficiency of precision spraying and mowing systems in orchards, vineyards, and other woody crop environments.

This study presents a comprehensive tomato harvesting robot system addressing three critical technical aspects: 1) optimal manipulator configuration design, 2) robust environmental recognition, and 3) efficient end effector control. For the manipulator configuration design, four different mounting configurations were systematically evaluated, with the vertical configuration featuring an offset end effector achieving the highest target reachability of 97.7%. For environmental recognition, a multi-sensor system that combines RGB and depth (RGBD) cameras and light detection and ranging (LiDAR) was implemented, utilizing depth filtering to suppress outliers. The end effector integrates suction and cutting mechanisms, employing a suction pad with conforming motion and Bowden cable-driven scissors. A bunch model was developed based on actual fruit bunches to create a testing environment with diversity and reproducibility. Field experiments conducted in a commercial greenhouse demonstrated continuous harvesting operations with a 68% suction success rate and a 45% overall harvesting success rate across 159 target fruits from 200 bunches. Additionally, the fruit position distribution in the field was measured, which can be utilized for layout optimization. This study contributes to advancing practical agricultural robotics by providing validated solutions for the three fundamental challenges in robotic crop manipulation.

The increasing demand for agricultural productivity and labor efficiency has driven the adoption of smart farming technologies, with global precision agriculture expected to grow at over 13% annually and labor shortages affecting nearly 40% of farming operations in developing regions. Additionally, inefficient traditional practices contribute to reduced yield and increased resource wastage, emphasizing the need for automation in agriculture. These environments demand intelligent, multifunctional machines capable of performing continuous field operations with accuracy and consistency. Traditional farming methods rely heavily on manual labor and separate machinery for each task, leading to increased time consumption, higher costs, inconsistent seed placement, and inefficient water usage. Furthermore, conventional systems lack real-time monitoring and remote-control capabilities, reducing overall productivity and adaptability to changing conditions. To address these challenges, the proposed IoT Agricultural Robot for Automatic Ploughing, Seeding, and Sprinkling utilizes the ATmega328 microcontroller integrated with an IoT module to develop a smart and automated farming solution. The robot is equipped with DC geared motors for soil ploughing, a seed dispensing mechanism for uniform seed placement, and an automated sprinkler system for irrigation. The IoT module enables remote control and real-time monitoring of the robot’s operations, allowing farmers to manage field activities efficiently. The system operates on a regulated power supply and is programmed using embedded C through the Arduino IDE, ensuring reliable performance. This integrated approach enhances agricultural efficiency, reduces labor dependency, optimizes resource utilization, and supports the development of sustainable and smart farming practices

Our global food system faces growing challenges such as population growth, climate change, resource constraints, and food loss. This set of threats has begun to erode the stability of food security efforts and challenge the long-term sustainability goals outlined by global organizations. To respond effectively, the sector needs concrete and forward-looking innovations that reflect the objectives of the Sustainable Development Goals (SDGs) of the United Nations (UN), especially the commitment in Goal 2 to eliminate hunger. In this study, we examine how agricultural robotics can support the shift toward more resilient and sustainable food systems, particularly in areas where classical methods are under strain. It brings together perspectives from technology, sustainability, and policy, aiming to bridge broad global priorities with everyday realities faced in local contexts. To structure the discussion in a concise way, our analysis is framed around five different, yet interrelated, dimensions. First, we use a crisis-framing perspective to explain why food system reform has become urgent and to show how these pressures align with key SDG priorities. The second dimension outlines a simple taxonomy that groups agricultural robots according to their domain and intended function while also highlighting ongoing technical issues such as interoperability. The next dimension examines how robotics is being amalgamated with precision farming tools, Internet of Things (IoT) platforms, artificial intelligence (AI), and big data systems. Collectively, these technologies facilitate more autonomous field operations and support faster, data-driven decision making. The sustainability dimension evaluates how these technologies affect environmental, economic, and social outcomes in the agricultural sector. This comprehensive review highlights several potential advantages, such as reduced chemical inputs, improved water efficiency, improvements in soil quality, more efficient use of labor, and new employment opportunities in rural and remote areas. In the final dimension, this study turns to global case studies, drawing comparative insights between developed nations such as Australia and the United States, and emerging economies including Brazil, India, and China. Across these diverse contexts, agricultural robotics consistently demonstrate the capacity to boost productivity, reduce waste, and make more efficient use of resources. It is apparent that these gains extend beyond the farm, contributing to environmental stewardship and broader socio-economic development. Yet, the path to widespread adoption is far from straightforward. Farmers and policymakers alike confront persistent barriers: the high upfront costs of robotic systems, gaps in technical expertise, difficulties in ensuring interoperability across platforms, and pressing ethical questions around data governance and automation. Overcoming these challenges is not simply a technical exercise; it is a prerequisite for realizing the full promise of robotics in reshaping global food systems for a more sustainable future.

Dataset of RGB images of healthy grapevine leaves and with downy mildew, powdery mildew, Esca complex, and erineum mite symptoms.