Innovations in Sensor-Based Systems and Sustainable Energy Solutions for Smart Agriculture: A Review

Abstract

Smart agriculture is transforming traditional farming by integrating advanced sensor-based systems, intelligent control technologies, and sustainable energy solutions to meet the growing global demand for food while reducing environmental impact. This review presents a comprehensive analysis of recent innovations in smart agriculture, focusing on the deployment of IoT-based sensors, wireless communication protocols, energy-harvesting methods, and automated irrigation and fertilization systems. Furthermore, the paper explores the role of artificial intelligence (AI), machine learning (ML), computer vision, and big data analytics in monitoring and managing key agricultural parameters such as crop health, pest and disease detection, soil conditions, and water usage. Special attention is given to decision-support systems, precision agriculture techniques, and the application of remote and proximal sensing technologies like hyperspectral imaging, thermal imaging, and NDVI-based indices. By evaluating the benefits, limitations, and emerging trends of these technologies, this review aims to provide insights into how smart agriculture can enhance productivity, resource efficiency, and sustainability in modern farming systems. The findings serve as a valuable reference for researchers, practitioners, and policymakers working towards sustainable agricultural innovation.

1. Introduction

1.1. Background and Motivation

The growing global population, increasing food demand, and depletion of natural resources present critical challenges for modern agriculture. Conventional farming practices often face limitations such as inefficient resource use, high dependency on manual labor, and inconsistent yield outcomes. To address these issues, smart agriculture has emerged as a promising approach by integrating advanced sensor-based systems, renewable energy solutions, and data-driven decision making. A significant technological transformation is occurring in agriculture through the adoption of Internet of Things (IoT), artificial intelligence (AI), computer vision, and big data analytics. Computer vision systems, particularly in livestock monitoring, crop surveillance, and pest detection, enable real-time observation and automated management of agricultural operations [1,2]. In recent years, the emergence of 5G communication technologies has further transformed smart agriculture by enabling ultra-low latency, high reliability, and massive IoT connectivity. The paper of Rehman [3] emphasizes the pivotal role of 5G in integrating remote-sensing data with real-time decision making, thereby enhancing productivity and sustainability in modern agricultural systems. Meanwhile, big data analytics empowers farmers by offering predictive insights into crop health, soil conditions, and irrigation needs, facilitating precision farming practices [4,5]. Despite remarkable technological advancements, there remains a gap in the holistic integration of RF energy-harvesting techniques, sustainable energy management, and precision agriculture systems [6]. The efficient and wide deployment of smart sensing systems is often constrained by power availability, especially in remote and rural farming environments. Therefore, designing energy-autonomous sensing networks powered by RF energy harvesting and sustainable energy sources becomes crucial. The main motivation of this review is to systematically analyze recent innovations in sensor-based smart agriculture technologies, critically discuss energy-harvesting solutions to power such systems, and explore sustainable energy strategies for developing resilient and efficient agricultural ecosystems. The review highlights the integration of sensing, communication, energy harvesting, and data analytics technologies and proposes future directions to address existing gaps in smart agriculture. This paper’s main contribution is to provide a comprehensive, updated review combining sensor technologies, wireless communication protocols, renewable energy integration, and advanced data analytics for smart agriculture. It specifically emphasizes the role of wideband RF energy harvesting and self-powered sensing systems—a relatively underexplored but essential field for next-generation agricultural applications.

Weather forecasting, soil nutrient monitoring, insect and pest control, early notifications of viruses and bacteria, automatic control systems, remote sensing, and various other parameters form the core components of smart agriculture. These technologies enhance the agroecological environment, increase production and quality, and reduce the use of pesticides and chemical fertilizers, thereby improving the reliability and sustainability of agricultural activities. Additionally, the adoption of renewable energy solutions such as solar and wind power further promotes sustainability by reducing dependence on fossil fuels and minimizing environmental impact.

This paper introduces the role of smart agriculture in providing food security for both developing and developed countries. For developed countries, the main concerns are optimizing large-scale farming and increasing productivity while lowering environmental impacts. In contrast, developing countries face challenges such as land scarcity and labor shortages. Deploying smart agriculture systems can significantly improve productivity, reduce hunger, and provide resilience against the challenges posed by climate change, making it essential for ensuring food security for the growing global population.

1.2. Objectives of the Study

The goal of this study is to systematically analyze recent innovations in sensor-based smart agriculture technologies, critically discuss energy-harvesting solutions to power such systems, and explore sustainable energy strategies for developing resilient and efficient agricultural ecosystems. The main objectives of this paper are to:

• Summarize the latest developments in smart agriculture technologies.
• Highlight the growing role of artificial intelligence and machine learning in agricultural processes.
• Discuss the benefits and limitations of various smart agriculture systems.
• Provide insights into future research directions and potential applications of these technologies.

1.3. Main Contributions

This paper makes several key contributions to the field of smart agriculture:

• Comprehensive Review: Provides an in-depth review of recent advancements in IoT-based sensors, wireless communication protocols, energy-harvesting methods, and automated irrigation and fertilization systems.
• AI and ML Integration: Emphasizes the growing role of artificial intelligence and machine learning in monitoring and managing various agricultural processes, including crop health assessment, pest control, and soil and water resource optimization.
• Technological Insights: Discusses the benefits and limitations of various smart agriculture systems, offering valuable insights for researchers and practitioners.
• Future Directions: Identifies potential future research directions and applications of smart agriculture technologies, contributing to the ongoing development of sustainable and efficient farming practices.

1.4. Various Wireless Nodes

One advanced technology associated with smart agriculture is remote sensing. Wireless sensor networks (WSNs) are a critical component of smart agriculture, enabling real-time monitoring and data collection. Various wireless nodes are used in agricultural applications to measure environmental parameters such as temperature, humidity, soil moisture, and light. To monitor the physical conditions of the land and environment, wireless sensor network (WSN) technology creates a network of connected wireless nodes. Wireless nodes in the agricultural sector are illustrated in Figure 1. This figure shows different types of wireless nodes, including MICA2, Cricket, IRIS, and MICAz, which are used to monitor environmental parameters such as temperature, humidity, and soil moisture. These nodes are essential for real-time data collection and decision making in smart agriculture.

By sensing air temperature, pressure, humidity, soil moisture, and other elements, the health of plants can be significantly improved. Wireless sensor networks (WSNs) play a crucial role in this process by providing real-time data on these parameters. These sensors help farmers make informed decisions about irrigation, fertilization, and pest control, thereby optimizing resource usage and enhancing crop yield. Additionally, the integration of WSN with IoT technologies allows for remote monitoring and automated control of agricultural processes, further improving efficiency and sustainability. Some of the widely used wireless nodes in agricultural practice are listed in

Table 1. Overview of widely used wireless nodes in the agricultural sector.

S/N	Wireless Node	Signaling Rate	Sensing Parameters	Reference
1	MICA2	38.4 K Baud	Sounder, video sensor, accelerometer, GPS	[7]
2	Cricket	38.4 K Baud	Temperature, light, pressure, humidity, relative humidity, acoustic, magnetometer	[8]
3	IRIS	250 Kbps	Light, pressure, acceleration, magnetic, relative humidity, acoustic, seismic, video sensor	[9]
4	MICAz	250 Kbps	Light, video sensor, GPS, relative humidity, humidity, magnetometer, temperature, pressure, accelerometer, acoustic, sounder, microphone	[9]
5	MICA2DOT	38.4 K Baud	GPS, relative humidity, light, temperature, humidity, pressure, accelerometer, acoustic	[7]
6	Imote2	250 Kbps	Temperature, light, accelerometer, humidity	[8]

. This table lists different wireless nodes, their signaling rates, and the sensing parameters they monitor. For example, the MICA2 node has a signaling rate of 38.4 K Baud and can sense temperature, light, pressure, and humidity, among other parameters.

1.5. Various Agricultural Sensors

Traditional agriculture systems often rely on human perception and experience rather than smart sensors, which can lead to inconsistent production quality. To address this issue, real-time sensing is crucial for continuously tracking environmental conditions and changes in plant health. This technology enables farmers to make timely decisions, allowing for early treatment, resource optimization, cost reduction, improved crop health, and increased efficiency. By minimizing the excess use of resources such as water, pesticides, and fertilizers, real-time sensing lowers the environmental impact and maximizes yield production through data analysis on plant growth and environmental conditions.

Table 2. Overview of various sensors used in smart agriculture. Adapted from ref. [10].

Serial No.	Sensor Name	Parameters
1	EC 250, ECH2O	Soil temperature, soil moisture, salinity level of water, conductivity
2	107-L, LT-2 M, 100K6A1B, MP406	Temperature of the plant
3	H2TM, 237 LWS	Level of CO2, H2 and Temperature, Wetness of the plant
4	CM1000TM, YSI 6025	Photosynthesis
5	LW100, TT4	Moisture, temperature, wetness of the plant
6	TPS-2	Photosynthesis and level of CO2
7	Cl-340, PTM-48A	Photosynthesis, moisture, temperature, wetness, H2, CO2 Level of the plant
8	CM-100, MSO- 70	Temperature, pressure and humidity of the air, wind speed
9	HMP45C, Cl-340, XFAM-115KPASR, SHT71, SHT75	Temperature, pressure and humidity of the air
10	107-L AT	Air temperature

summarizes the various sensors used in smart agriculture. This table provides details on different types of sensors, including soil moisture sensors, temperature sensors, and photosynthesis sensors, along with the specific parameters they measure. These sensors are essential for optimizing agricultural practices and improving crop productivity.

Integrating IoT and smart sensors for smart farming is illustrated in Figure 2. This figure illustrates different types of sensors, such as soil moisture sensors, temperature sensors, and photosynthesis sensors, which are deployed in smart agricultural systems. These sensors provide critical data for optimizing irrigation, fertilization, and pest control practices. These devices play a crucial role in monitoring and managing agricultural processes, providing real-time data on environmental conditions and plant health. By leveraging these sensors, farmers can optimize resource usage, improve crop yield, and enhance the overall efficiency and sustainability of their farming practices.

1.6. List of Wireless Communication Protocols (WCP)

Wireless communication protocols are crucial for the automatic control systems in smart agriculture. These protocols enable seamless communication between sensors, controllers, and data processing units. Technologies such as Cellular, ZigBee, 6LoWPAN, RFID, Bluetooth, Wi-Fi, GPTS, SigFox, and LoRaWAN are commonly used in this domain [11,12,13,14,15,16,17,18,19,20,21,22,23]. These platforms connect with sensors to collect information and transmit data to farmers, reducing the need for physical presence in the field and enabling real-time adjustments as needed. Wireless communication systems eliminate the need for physical cables or communication lines, thereby reducing the need for expensive infrastructure.

Table 3. List of wireless communication protocols (WCP) used in the smart agricultural sector. Adapted from ref. [10].

Communication Protocols	Network Topology	Data Rate	Standard	Power Consumption	Communication Range
6LoWPAN Technology	Star, Mesh	0.3–50 Kbps	IEEE 802.15.4 [24]	Low	2 to 5 km urban, 15 km sub-urban
ZigBee Technology	Star, Mesh, cluster	250 Kbps	IEEE 802.15.4 [24]	Low	10 to 100 m 15 km sub-urban
Bluetooth Technology	Star, Bus	1–2 Mbps	IEEE 802.15.1 [25]	Low	30 m
RFID Technology	P2P	50 tags/s	RFID [26]	Ultra Low	10 to 20 cm
LoRa WAN Technology	P2P, Star	27–50 Kbps	IEEE 802.11ah [27]	Very Low	5 to 10 km
Wi-Fi Technology	Star	1–54 Mbps	IEEE 802.11 [28]	Medium	50 m

provides an overview of some commonly used wireless communication protocols (WCP) in the smart agricultural sector [10]. This table outlines various communication protocols, their network topology, data rate, standard, power consumption, and communication range. These protocols facilitate efficient data transmission and remote monitoring in smart agriculture.

An overall smart agricultural system is presented in Figure 3. This figure demonstrates how various components, including IoT-based sensors, wireless communication protocols, and automated control systems, work together to enhance agricultural productivity and sustainability.

1.7. Structure of the Paper

The paper is organized as follows:

Section 2: Provides a detailed review of IoT applications in agriculture, including precision irrigation, soil and crop monitoring, livestock health tracking, disease prevention, and automated fertilization. Section 3: Discusses the parameters associated with smart agriculture and their relevance to IoT applications. Section 4: Presents the identification of harmful insects and diseases, highlighting the role of IoT in pest control. Section 5: Describes additional systems associated with monitoring and control in smart agriculture. Section 6: Explores energy-harvesting approaches in agriculture, focusing on sustainable energy solutions. Section 7: Discusses the differences in methods for field and greenhouse cultivation, emphasizing the use of stationary sensors in greenhouses and mobile platforms in fields. Section 8: Outlines future directions for research and development in smart agriculture, focusing on the integration of AI and machine learning, expansion of renewable energy sources, and development of affordable, scalable technologies. Section 9: Summarizes the main findings and contributions of the paper, highlighting the importance of continued innovation and research in smart agriculture to achieve sustainable and efficient farming practices.

2. IoT Applications in Agriculture

2.1. Overview of IoT in Agriculture

The Internet of Things (IoT) has revolutionized agriculture by enabling real-time monitoring and control of various farming operations. IoT applications in agriculture leverage interconnected sensors, devices, and data analytics to optimize resource usage, improve crop yields, and enhance sustainability. This section provides a detailed review of key IoT applications in agriculture, including smart irrigation systems, smart fertilization systems, soil and crop monitoring, livestock health tracking, and early disease detection and pest control systems.

Table 4. List of IoT applications and their corresponding parameters.

Serial No.	Parameter	IoT Application	Ref.
1	Soil moisture	Precision irrigation	[29]
2	Temperature	Soil and crop monitoring	[30]
3	Nutrient levels	Automated fertilization	[6]
4	Livestock health	Livestock health tracking	[5]
5	Environmental conditions	Disease prevention	[31]

links various parameters to their corresponding IoT applications in agriculture. It provides a clear overview of how different parameters are monitored and managed using IoT technologies, along with references to studies where these applications have been studied.

2.2. Smart Irrigation System

The smart irrigation system is a modern technology designed to optimize water usage by continuously monitoring soil moisture, weather conditions, crop needs, and other environmental factors. This system minimizes water waste and ensures that plants receive the precise amount of water they need, thereby increasing sustainability and yield production. These systems automatically adjust irrigation schedules to ensure optimal water usage, reducing waste and improving crop yields. IoT-based precision irrigation has been shown to reduce water consumption by up to 30% while enhancing crop productivity. Recent advancements include the integration of AI algorithms for predictive analytics and remote-sensing technologies for real-time monitoring [29]. Figure 4 shows a conceptual model of a smart irrigation system [32]. This figure shows how smart irrigation systems use sensors to monitor soil moisture, weather conditions, and crop needs, ensuring precise water application and minimizing waste.

In many underdeveloped areas of the world, such as Africa, irrigation is often manually operated. Introducing smart irrigation systems in these regions can address water scarcity issues [33]. The system automatically optimizes water usage by controlling water pumps based on reservoir capacity and soil moisture levels, ensuring water is delivered precisely where it is needed [34].

Recent concerns about food limitations in India due to its rapidly increasing population highlight challenges such as irregular rainfall and water scarcity [35]. A proposed technique involves observing changes in air temperature and humidity, sending signals to a microcontroller that opens and closes valves based on moisture sensor signals to ensure water is supplied directly to plant roots. A buzzer system alerts users when irrigation is needed, streamlining the process of managing water supply to crops. This smart technology ensures optimal water usage and makes agriculture more efficient and sustainable [36,37]. The authors [38] provide a comprehensive overview of smart irrigation techniques utilizing AI-based predictive analytics and remote soil moisture sensing. Their research demonstrated a 30% increase in water use efficiency by combining real-time sensor data with machine-learning models for adaptive irrigation control.

A sensor-based automated irrigation system utilizing the latest IoT-based technology to enhance agricultural efficiency was proposed in [39]. It introduced microcontrollers named ATMEGA328P to monitor soil moisture levels and control precise water application, preventing inefficient irrigation processes. Automated systems reduce human errors, increase energy efficiency, yield production, and profitability. This paper also highlights the growing demand for such technologies in various agricultural sectors, emphasizing their role in sustainable and smart irrigation systems [39].

2.3. Smart Fertilization System

A smart fertilization system is a modern agricultural technique designed to optimize the use of fertilizers in the field. Similar to smart irrigation, it works with sensors to analyze data and automatically make decisions to ensure that plants receive the necessary nutrients in a timely manner while reducing the environmental impact of fertilizer waste. Figure 5 shows a diagram of a smart fertilization system. This figure illustrates how smart fertilization systems use sensors to analyze soil nutrient levels and environmental conditions, enabling automated and optimized fertilizer application to improve crop health and reduce environmental impact.

The fertilization decision is supported by integrating three key models: the function of soil fertilizer effect, dissimilar subtraction method, and soil nutrient balance. These models collectively form a comprehensive decision-making system [41]. Statistical analysis minimizes errors through real-time data acquisition from sensing elements, and GIS (Graphical Information System) aids in generating accurate soil fertility maps and fertilizer plans [42]. The system follows a decision support system (DSS) framework with data, model, and dialogue subsystems for flexible data access. Based on current weather and soil conditions, data are implemented on a web-based platform (ASP.NET and SQL 2000), allowing real-time communication and dynamic updates. Early data-driven fertilization decisions improve agricultural productivity [42,43].

The authors [44] proposed deploying solar-powered soil moisture, nutrient, and temperature sensors across farmlands, interconnected through wireless IoT networks. Real-time data transmission allows AI-driven platforms to dynamically optimize irrigation and fertilization schedules, minimizing resource wastage. The renewable energy-based model ensures energy self-sufficiency, particularly in off-grid and remote agricultural regions, supporting the broader vision of sustainable smart farming.

2.4. Soil and Crop Monitoring

IoT sensors provide valuable data on soil moisture, temperature, nutrient levels, and crop health. This information helps farmers make informed decisions about irrigation, fertilization, and pest control. IoT-based soil and crop monitoring systems enable precise management of agricultural resources, leading to improved crop yields and reduced environmental impact. Recent studies [30] have demonstrated the effectiveness of IoT sensors in optimizing soil and crop management practices.

2.5. Livestock Health Tracking

IoT devices are used to monitor livestock health, track movement, and detect diseases early. Wearable sensors and RFID tags provide real-time data on animal behavior, body temperature, and heart rate. This information helps farmers identify health issues promptly and take preventive measures to improve animal welfare. IoT-based livestock health tracking systems have been shown to enhance productivity and reduce veterinary costs [31].

2.6. Early Disease Detection and Pest Control System

Early disease detection and pest control systems are technological tools used to identify harmful insects and a range of diseases. These systems continuously monitor crop health using a variety of sensors and automation, acting on data before pests or diseases can seriously damage the crops. This method improves sustainable farming practices and minimizes harm from diseases and insects. Figure 6 shows photos of some harmful pests [45].

The significant impact that plant diseases have on crop production has been studied extensively [46,47,48,49,50]. Techniques for identifying plant diseases include enzyme-linked immunosorbent assays (ELISA) and polymerase chain reactions (PCR). These methods involve thorough collection and processing, which can cause delays in disease identification. New technologies apply various techniques that can be integrated into sensor systems, such as spectroscopic techniques for imaging and volatile organic compound (VOC) profiling for disease detection.

Table 5. Recent studies on plant disease detection using molecular techniques.

Plant/Crop	Pathogen	Type	Molecular Method(s)	Reference
Tomato	Fusarium sambucinum	Fungi	Quantitative PCR (qPCR)	[52]
Maize	Fusarium spp. and Magnaporthiopsis maydis	Fungi	Real-time PCR targeting ITS region	[53]
Olive	Fungal communities causing leaf spots	Fungi	PCR-based identification	[54]
Potato/tomato	Phytophthora infestans	Oomycete	Loop-mediated isothermal amplification (LAMP)	[55]
Tomato	Pseudomonas syringae pv. tomato	Bacteria	LAMP targeting hrpZ gene	[56]
Various crops	Plant viruses	Virus	PCR, ELISA, next-generation sequencing (NGS)	[57]

summarizes recent studies on plant disease detection using molecular techniques. It highlights various detection methods such as ELISA, PCR, quantitative PCR (qPCR), loop-mediated isothermal amplification (LAMP), and next-generation sequencing (NGS), applied to a range of plant-pathogen interactions. These techniques play a crucial role in early disease identification, which is essential for effective crop protection and sustainable agricultural management. The authors [51] extensively reviewed AI and automation techniques, highlighting the application of all-terrain vehicles integrated with multispectral imaging sensors to achieve early-stage disease detection, thereby improving intervention strategies and minimizing crop losses.

2.7. Decision-Making Systems

Decision-making systems in smart agriculture are essential for optimizing various agricultural processes such as irrigation, fertilization, and pest control. These systems leverage data from IoT sensors, AI algorithms, and machine-learning models to make informed decisions. While machine-learning-based models are widely discussed, mechanistic models also play a crucial role in decision-making systems. Mechanistic models are based on the physical and biological processes governing the behavior of agricultural systems. By integrating mechanistic models with machine-learning-based models, decision-making systems can achieve higher accuracy and reliability in managing fertilization, irrigation, and pest control. Recent studies have demonstrated the effectiveness of hybrid mechanistic and machine-learning models in optimizing agricultural practices and improving crop yield [52].

3. Some Parameters Associated with Smart Agriculture

The application of sensor-based technology to optimize agricultural processes is the main concern of this paper. Several essential parameters, such as soil moisture, temperature, pH, nutrient levels, and CO₂ concentration, are continuously monitored in real-time using various sensors to improve crop productivity. By applying these technologies, research aims to increase crop productivity, improve sustainable farming methods, and enhance resource efficiency in environments with limited resources.

3.1. Soil Moisture

Soil moisture refers to the amount of water content mingled with the soil. It can be expressed in terms of volume or weight and plays a crucial role in plant growth and agricultural productivity. Soil moisture affects factors such as soil temperature, plant water availability, and nutrient availability.

A ground penetrating radar (GPR) methodology to measure soil moisture was introduced in [58]. Widely used soil moisture sensors in agricultural settings include neutron-scattered probes, which use radiation scatter techniques to detect soil moisture by calculating fluctuations in flux density due to the moisture content of the soil. Specific licenses are required for using these sensors. Studies in [59,60] examine the accuracy of soil water content measurement devices, focusing on neutron scattering (NS) and capacitance probe (CP) gauges in measuring water content in different soil horizons at various sites, including Amarillo and Big Spring, Texas. These studies involve comparing NS and CP gauges [61].

3.2. Soil pH Level

Soil pH is a method of measuring the acidity or alkalinity of the soil. It is a key parameter that can be used to make informative analyses regarding soil characteristics. Soil pH can be measured both quantitatively and qualitatively. If the soil is acidic, the pH level is below 7; if the soil is alkaline, the pH level is above 7. Soil pH affects nutrient availability, microbial activity, and soil health, making it essential for the agricultural sector.

Investigations have used different methods for soil pH estimation, focusing on the Varies pH Manager, which uses ion-selective antimony electrodes to measure real-time, accurate pH data [60,62]. To improve accuracy, field-specific calibration is performed by adjusting measurements using reference lab data. The study also discusses sampling strategies, such as point composite sampling on a grid, and integrates soil apparent electrical conductivity (ECa) as a co-variable. Validation against standard sampling methods has demonstrated a more accurate representation of soil pH.

3.3. Yield Monitoring

Yield monitoring refers to the process of measuring and analyzing the quantity and quality of crops harvested from a given area over a specific term. Yield monitoring is essential in modern agricultural systems as it provides an overall view of production data, which is used to improve and optimize farming practices. Yield monitoring allows farmers to identify patterns, variabilities, and factors affecting crop production.

Some studies have employed various methods to measure spatial and temporal changes, covering several counties from 1982 to 2013, to estimate county-level yields. Precipitation use efficiency (PUE) is calculated by dividing annual and growing season yields by corresponding precipitation to determine water use efficiency. These methods provide a brief description of crop yield and the productive use of water for future agricultural planning [63,64].

3.4. Sensing of the Weather and Environment

Weather and environment sensing involves monitoring atmospheric and environmental quantities in real-time. These data are essential for weather forecasting, environmental management, agriculture, disaster preparedness, and climate research. By providing accurate and timely information on weather conditions, these sensing systems help farmers make informed decisions about planting, irrigation, and harvesting, thereby optimizing agricultural practices and improving crop yields. Additionally, real-time environmental data support effective disaster preparedness and response, mitigating the impacts of extreme weather events on agricultural productivity.

3.5. Sensing Micronutrients in Soil

The process of detecting and measuring the vital nutrients needed for plants to grow, such as potassium (K), phosphorus (P), and nitrogen (N), is known as macronutrient sensing in soil. Crop productivity depends on these macronutrients, and precise monitoring helps maximize fertilization and resource utilization in agriculture. To enhance real-time nutrient analysis and achieve high accuracy in measurement, automated soil sampling systems have been developed [65,66,67]. Field test results are promising, but mechanical issues and the need for improved extraction methods remain challenges.

3.6. Remote Sensing for Smart Agriculture

Remote-sensing technology for smart agriculture uses satellite, drone, and aircraft-based sensors to gather real-time data on crop health conditions, soil moisture, water usage, and weather conditions. This technology helps farmers make early decisions based on real-time data, thereby improving agricultural efficiency, increasing yields, and promoting sustainable practices that allow farmers to optimize resources.

Different types of sensors, such as stationary and mobile sensors, measure various characteristics of crops in real-time [68,69,70,71,72]. These sensors are mainly designed to measure weed densities, crop height, leaf reflectance, and moisture status, which are essential for making informed decisions. Advanced techniques such as light detection and ranging (LiDAR) and fluorescence spectroscopy are also employed. Additionally, these studies discuss the adequate nitrogen levels needed to calibrate the sensors. These methods collectively contribute to advancements in agriculture and crop management [73].

Remote and proximal sensing technologies are essential components of precision agriculture. Hyperspectral, multispectral, and RGB imaging provide detailed information on crop health, soil conditions, and water usage. Measurements of sunlight-induced fluorescence and thermal imaging offer insights into photosynthetic processes and plant stress. Reflectance indices, such as NDVI, PRI, WI, and NDWI, are valuable tools for assessing plant health, photosynthetic activity, and water content. These technologies enable precise monitoring and management of agricultural systems, leading to improved productivity and sustainability [74,75]. Comparison among various imaging techniques used in smart agriculture is given in

Table 6. Comparison of imaging techniques based on spectral range, resolution, applications, advantages, and limitations.

Imaging Technique	Spectral Range	Resolution	Key Applications	Advantages	Limitations	Ref.
Hyperspectral imaging	400–2500 nm	<1 nm	Crop health monitoring, soil analysis, water usage	Detailed spectral information, precise monitoring	High cost, complex data processing	[76,77]
Multispectral imaging	400–1000 nm	10–20 nm	Yield prediction, crop monitoring, field robotics	Cost-effective, rich spectral information	Lower spectral resolution compared to hyperspectral	[78,79]
RGB imaging	400–700 nm	1–2 nm	Disease detection, plant growth assessment	Cost-effective, widely used	Limited spectral information	[80,81]
Sunlight-induced fluorescence	650–800 nm	<0.5 nm	Photosynthetic activity monitoring, plant health	Accurate photosynthesis estimation, ecosystem monitoring	Requires high spectral resolution	[82,83]
Thermal imaging	8–14 µm	0.1–0.5 °C	Crop stress detection, irrigation management	Real-time monitoring, detects stress and malfunctions	Sensitive to environmental conditions	[84,85]

.

3.6.1. Hyperspectral Imaging

Hyperspectral imaging is an advanced technology that captures detailed spectral information across a wide range of wavelengths. This technology is particularly useful in smart agriculture for monitoring crop health, soil conditions, and water usage. Hyperspectral imaging allows for the identification of specific chemical and physical properties of plants, enabling precise monitoring and management of agricultural systems. Recent advancements in hyperspectral imaging include the development of mini-sized and low-cost airborne sensors, such as Headwall Micro-Hyperspec and Cubert UHD 185-Firefly, which have made this technology more accessible to agricultural applications. Hyperspectral imaging is becoming increasingly important for mapping crop biophysical and biochemical properties, soil characteristics, and crop classification [76,77].

3.6.2. Multispectral Imaging

Multispectral imaging uses sensors capable of detecting spectral information across various wavelength ranges to acquire multi-channel data. This technology is widely applied in agriculture for collecting crop information and predicting yield. Multispectral imaging provides valuable support for applications such as image segmentation, crop monitoring, field robotics, and yield estimation. Recent studies have focused on the use of multispectral technology for plant yield prediction, highlighting key areas such as chlorophyll content, remote sensing, convolutional neural networks (CNNs), and machine learning. Multispectral imaging enables researchers to collect comprehensive biological information about observed objects or areas, facilitating a deeper understanding of plant growth mechanisms and ecosystem function [78,79].

3.6.3. RGB Imaging

RGB imaging, which captures images in red, green, and blue channels, is a cost-effective and widely used technology in smart agriculture. This imaging technique is valuable for monitoring crop health, detecting diseases, and assessing plant growth. RGB imaging can be integrated with deep-learning algorithms to enhance its capabilities, allowing for accurate and real-time analysis of agricultural data. Recent advancements in RGB imaging include the use of UAV-based platforms for large-scale crop monitoring and the development of deep-learning models for plant disease detection. These innovations have significantly improved the efficiency and accuracy of RGB imaging in agricultural applications [80,81].

3.6.4. Measurements of Sunlight-Induced Fluorescence

Sunlight-induced fluorescence (SIF) is a promising proxy for monitoring photosynthetic activity and plant health. SIF measurements provide insights into the actual photosynthesis and plant status, making it a valuable tool for smart agriculture. Recent advancements in SIF technology include the development of portable field spectroradiometers and the launch of satellite missions such as the Earth Explorer 8 FLuorescence EXplorer (FLEX) mission. These technologies enable accurate estimation of SIF from both ground and space platforms, facilitating large-scale monitoring of canopy fluorescence at the ecosystem level. SIF measurements are crucial for calibrating and validating satellite data, ensuring reliable monitoring of plant health and productivity [82,83].

3.6.5. Thermal Imaging

Thermal imaging is a powerful tool for monitoring crop health, detecting stress, and identifying irrigation malfunctions. This technology captures temperature variations in plants, providing valuable data for optimizing irrigation schedules, detecting pests and diseases, and improving crop management strategies. Recent advancements in thermal imaging include the integration of machine-learning algorithms to enhance its capabilities. Thermal imaging has been successfully used to identify malfunctions in subsurface drip irrigation systems and to monitor crop stress in various agricultural settings. These innovations have revolutionized agricultural practices by providing real-time, data-driven insights into crop health and productivity [84,85].

4. Identification of Harmful Insects and Diseases

Harmful insect identification is a crucial aspect of pest management in agriculture. Accurately detecting and identifying harmful insects helps farmers take the necessary action to prevent damage to crops. Modern technology and traditional identification methods allow farmers and researchers to detect pest infestations early. The classification of disease-detection techniques is shown in Figure 7. This figure categorizes various methods used for detecting plant diseases, including enzyme-linked immunosorbent assays (ELISA), polymerase chain reactions (PCR), and spectroscopic techniques. These methods are crucial for early disease detection and effective crop management.

Various innovative methods for improving pest detection and management in agriculture have been discussed in the literature. For example, an improved YOLOv3 model and image augmentation techniques have been used to enhance pest detection in maize [86]. Deep convolutional neural networks (DCNN) and transfer learning were employed for pest classification, fine-tuning the ResNet-50 model to improve accuracy. A dataset of nine common maize pests was used to train the model, incorporating various image augmentation techniques, particularly image cropping. A review on integrated pest management (IPM) proposes agroecological crop protection (ACP) as a viable alternative, providing a review of its evolution, strengths, and weaknesses, and examining the socio-economic, environmental, and health challenges [87].

The method of deep convolutional neural networks (DCNN) has been applied to achieve effective pest classification. Primarily, DCNN was used for pest clarification and recognition. The DCNN architecture, including ResNet-50, was used to classify ten types of pests. An image pyramid structure was employed to fuse features from different levels, and the fine-tuned pre-trained model on their specific pest dataset allowed for improved accuracy [88].

YOLOv3 technology has also been used to detect tomato diseases and pests. This improvement focused on multi-scale feature detection, allowing the model to better identify objects. An image pyramid structure was used to fuse features from different levels, helping to obtain feature maps of various scales, which were crucial for the accurate location of pests and diseases. The study collected image data of tomato diseases and pests in real natural environments, making the training data more effective [89].

Several object-detection models used to detect pest insects have focused on state-of-the-art (SoA) object models based on conventional neural networks. These papers reported back propagation-based visualization techniques that helped understand which insect features were considered by the models, allowing for qualitative validation of the algorithm’s decisions [90,91].

4.1. Crops Contaminated by Insects

Contaminated crops are those that are affected and damaged by harmful insects feeding on, infesting, or transmitting diseases to the plants. Insect contamination can significantly reduce crop production, leading to economic losses for farmers. The extent of insect contamination depends on the type of insect species, the stage of crop growth, and environmental factors.

Three deep-learning ensemble models for improving plant disease and pest detection are applied in [92,93,94]. The first model is a simple ensemble averaging method that combines multiple predictions to increase accuracy. In contrast, the second model, Early Fusion CNN Ensembles, integrates features before classification using a support vector machine (SVM). The third model, majority voting CNNs, predicts the majority vote from CNN models and also uses SVM. Six pre-trained CNNs, including AlexNet, GoogleNet, and ResNet, were fine-tuned and tested on a Turkey-Plant dataset, with majority voting achieving the highest accuracy (97.56%).

An artificial neural network (ANN) for identifying greenhouse pests through high-resolution image processing was introduced in [95,96]. The authors explored the use of CNN models for disease detection in tomato plants, focusing on two models: F-CNN, which processes full leaf images, and S-CNN, which focuses on segmented areas of disease symptoms. The models were trained using Python (version 3.8) and the TensorFlow deep-learning library, with images sourced from villages, farms, and plants. Brightness and blur adjustments were used for data augmentation to enhance dataset diversity. S-CNN achieved higher accuracy (98.6%) than F-CNN.

4.2. Crop Inspection

Crop inspection involves the systematic practice of crop fields to assess plant health, identify pests and diseases, and evaluate overall crop performance. Crop inspection is essential for farmers to protect yields and make informed decisions about crop management. Regular crop inspection allows farmers to identify potential problems early, enabling timely interventions before significant damage occurs.

Using high-definition images, an improved cascade R-CNN for detecting small pests was explored in [97]. The convolution block attention module (CBAM) model was integrated into the training, enhancing feature extraction based on shape and color. Atrous spatial pyramid pooling (ASPP) and feature pyramid networks (FPN) further improved feature extraction and fusion across different scales. These methods collectively achieved an 88.78% detection accuracy for citrus psyllids. An approach for detecting plant diseases and insect pests in honey pomelo leaves using CNN involved robots picking 510 random leaves, and the CNN model with seven layers was used to extract features like color and spots on the leaves in [98]. Compared to support vector machines (SVM), the CNN’s performance was more accurate. The study highlights the potential of deep learning to enhance pest and disease detection for robust farming solutions.

The authors of [99] discussed an extensive approach to improve plant pest surveillance on grape phylloxera in vineyards. Unmanned aerial vehicles (UAVs) equipped with RGB cameras were used to collect high-resolution imagery, which was processed through various techniques. Photogrammetry was applied to create digital elevation models, and a digital vigor model (DVM) was used to analyze vineyard conditions. The study also aimed to improve pest detection in vineyards by developing new hyperspectral vegetation indices and developing a predictive model utilizing ground-based and airborne data.

5. Some Additional Systems Associated with Monitoring and Control

Smart agriculture allows farmers to automate processes, optimize resources, and improve overall productivity. These systems integrate sensors, actuators, data analytics, and automation technologies to monitor real-time environmental conditions and control agricultural operations. Moreover, integrating these systems with cloud-based platforms can enable remote monitoring and control, providing farmers with real-time insights and the ability to manage their operations from anywhere.

A low-cost temperature hysteresis control system for agricultural greenhouses, using the ADT14 integrated circuit, was discussed in [100,101]. The designed system allowed users to program up to four temperature control points with a control range from −40 °C to +125 °C. It incorporated voltage reference and temperature sensing mechanisms for accurate temperature detection. Additionally, a relay control mechanism for automatic temperature adjustments was used. The simplicity of design, cost-effectiveness, and ease of installation made it ideal for small-scale greenhouses.

A cost-effective and easily controlled method for a wireless control system of agricultural motors was designed in [102]. By using mobile phones as remote controls, farmers were able to send SMS commands to start or stop the motor. An AT89C52 microcontroller was used to sense and monitor parameters such as overload, dry run, temperature, and water levels, and process these commands. The system also employed a timer functionality to ensure accurate responses, and in case of malfunctions or security issues, it provided SMS alerts to the farmers.

The authors of [103] designed and evaluated a steering control system for agricultural tractors. They used ultrasonic pulse-echo ranging, where ultrasonic transducers detected changes in ground height by measuring the time it took for acoustic pulses to return. Transverse array scanning with fixed ultrasonic transducers was used to scan the ground with higher sampling rates. The system was tested with mathematical modeling, stability criteria, and error approximation on a track treadmill where the tractor was tethered longitudinally but free to move laterally. These approaches ensured the effectiveness of the proposed system in agricultural applications.

6. Energy-Harvesting Approaches in Agriculture

This section focuses on utilizing energy by capturing sustainable energy sources such as solar panels, wind turbines, and biomass energy. These approaches play a vital role in reducing dependency on fossil fuels and improving the overall sustainability of farming operations. Various techniques have been developed to harness these renewable energy sources effectively in agricultural settings. Some of these techniques include the use of advanced sensor networks, innovative energy storage solutions, and efficient energy conversion methods. These energy-harvesting schemes are crucial for enhancing the efficiency and sustainability of modern agricultural practices. Furthermore, combining multiple renewable energy sources can create a more resilient and reliable energy supply for agricultural systems, ensuring continuous operation even in varying environmental conditions. Mushtaq et al. The authors [104] explored hybrid energy-harvesting techniques combining solar, RF, and piezoelectric sources, enabling uninterrupted power supply for distributed smart agriculture sensor nodes, even in remote locations. Some energy-harvesting schemes in agricultural systems based on wireless sensor networks (WSN) are illustrated in Figure 8. This figure illustrates various energy-harvesting methods, such as solar cells, wind turbines, piezoelectric converters, and microbial fuel cells, which are employed to power smart agricultural technologies and enhance sustainability.

The authors [105] investigated the integration of AI, IoT, and renewable energy-harvesting technologies to revolutionize agricultural machinery. They proposed embedding multisensor arrays within tractors, harvesters, and drones to monitor operational parameters such as temperature, fuel consumption, and mechanical stress. These sensors, partially powered by onboard solar panels and kinetic energy harvesters, enable autonomous data collection and predictive maintenance using AI-driven models. IoT connectivity ensures real-time transmission of data to cloud servers, where machine-learning algorithms optimize machinery performance dynamically. This synergistic system significantly reduces operational costs, minimizes downtime, and promotes sustainable, low-emission farming practices.

These energy-harvesting techniques are crucial for maximizing resource efficiency and reducing environmental impact. A summary of various energy-harvesting methods used in agricultural applications can be found in

Table 7. Approaches towards energy harvesting in agriculture.

Techniques	Protocol/Device	Energy/Power Harvesting	Sensors	Ref.
Solar cell	ZigBee (XBee-Pro S2)	240 mW	Temperature of air, soil moisture	[22]
	RFD 900	1.75 to 3 W	Concentration of CH4 and CO2 Concentration	[106]
	ZigBee (CC2530)	500 mW	Humidity of air, temperature of air, shadow detection	[107]
	IEEE 802.15.4	2 W	pH level, wind direction, temperature of air, humidity of air, wind speed	[108]
	C1110 RF module	500 mW	Humidity of air, camera, temperature of air, leaf wetness	[109]
	IEEE 802.15.4	1 W	Temperature, condensation system, leaf wetness, rain gauge	[110]
	ZigBee (Mica2 motes)	20 W	Humidity of air, temperature of air, soil moisture, soil temperature	[111]
Inductive coupling	ZigBee	2.4 W	Temperature of air, vibration, pressure, soil moisture	[112]
Magnetic resonant coupling	Not specified	1315 J	TProcessing of water, environments sensors for agriculture	[113]
Wind turbine	Zigbee	70–100 mW	Ambient temperature of air, rainfall, soil moisture	[114]
Piezoelectric convertors	ZigBee (CC2500)	200 µW	Vibration sensor	[115]
Water flow	ZigBee	16–19 mW	Soil moisture, relative humidity, irrigation control	[116]
Microbial fuel cell	LoRa	296 µW	Irrigation system	[117]

. This table lists various energy-harvesting techniques, the devices used, the energy/power harvested, and the sensors involved. For example, solar cells paired with ZigBee (XBee-Pro S2) can harvest 240 mW of power to monitor air temperature and soil moisture. These methods contribute to sustainable energy solutions in smart agriculture.

6.1. Solar Energy

Solar energy is a renewable and sustainable source of power that plays a crucial role in the global transition to sustainable energy systems. It is an abundant resource that can be converted into usable forms of energy. In recent decades, several advancements have been made to integrate solar energy into smart agriculture systems [118]. Researchers have worked on solar photovoltaic (PV) systems, developed solar pumps for dry-land areas, utilized microcontroller-based systems, and improved the efficiency of these systems.

The authors [107,119,120] explored the feasibility and impacts of collocating solar photovoltaic (PV) systems with aloe vera gel production using several key methods. A life cycle analysis (LCA) evaluated the economic and environmental impacts of both solar PV and aloe vera systems, covering all stages from manufacturing to recycling for solar PV and from cultivation to processing for aloe vera. A Monte Carlo simulation was used to assess uncertainties in parameters such as solar panel efficiency and aloe vera leaf-to-gel conversion rates. The economic analysis compared landowner returns from single-use and collocated land use on a 5-hectare plot, considering cases of grid-tied and off-grid. These methods efficiently assessed land and water in dry-land areas like northwestern India, providing insights into the viability of combining these two land uses.

Focusing on improving sustainability and efficiency, several advanced methods for solar-powered irrigation in agriculture were discussed in [121]. Solar pump development is a key area with innovations such as the collaboration between IBC SOLAR and Siemens to replace diesel engines with solar-powered systems, making irrigation more cost-effective in arid regions. Prototype testing in Namibia demonstrated the practical, low-cost integration of solar systems into the existing infrastructure. System integration explored how photovoltaic systems and IBC pump controllers seamlessly replaced diesel engines, ensuring minimal disruption to current irrigation setups. The emphasis on drip irrigation highlights its water-saving advantages, which are particularly crucial for regions with water scarcity as it maximizes crop yields with targeted water delivery. The paper also covered mobile solar energy systems such as the Wien Energy system, which offered mobility and remote monitoring via a smartphone app, further enhancing water management flexibility.

Some methods for developing and testing a solar-powered dry-land micro-irrigation agricultural system in India were introduced in [122]. A gravity-fed micro-irrigation system was integrated with a low-cost solar pump designed based on hydraulic studies to ensure efficient water distribution. Field experiments were conducted at the Central Research Institute for Dry-land Agriculture in Hyderabad, representing typical semi-arid conditions. Performance evaluation included measuring emitter flow variation (18.96%), Christiansen uniformity coefficient (93.65%), and distribution uniformity (91.55%), all of which indicated high system efficiency. Operational guidelines suggested using upper pond water to prevent clogging and operating the system during optimal hours. The system’s design was adaptable to different farm sizes, with recommendations for adjusting the pumping system and solar power generation. These methods aimed to provide a sustainable irrigation solution for smallholder farmers in water-scarce regions.

Several innovative methods to enhance solar-powered wireless sensor networks (WSNs) through adaptive energy management were presented in [123]. A key approach is the packet transmission control policy, which adjusts transmission periods based on predicted solar energy availability to maintain network performance. The solar energy prediction method improved prediction accuracy by 6.92% using cloudiness and solar radiation data. Adaptive control of packet transmission (ACSE) dynamically adjusted transmission periods by considering battery energy and harvested energy, optimizing energy use, and minimizing deadline misses. Simulations in MATLAB (version R2021a) demonstrated ACSE’s advantages over other strategies like greedy and lazy scheduling. The paper also addressed handling aperiodic tasks using a Poisson distribution, ensuring efficient management of irregular packet generation. Additionally, error analysis of solar energy predictions, particularly under cloudy conditions, suggested mitigation through long-term data collection. These methods improved the operational efficiency and reliability of WSNs powered by solar energy.

Several key methods for developing a solar-powered greenhouse wireless sensor network monitoring system were introduced in [110,124]. Microcontroller selection is critical, using the low-power MSP430F149 to extend battery life. Sensor integration included the DS18B20 digital temperature sensor, which was energy-efficient and operated with minimal wiring via a 1-wire bus. Energy management was achieved by powering the sensor nodes through solar energy, ensuring continuous operation without external power sources. The system employed wireless communication for data transmission using low-power components like the nRF24L01 chip to minimize energy consumption. Additionally, the system allowed for data monitoring and control of environmental parameters like temperature and humidity, reducing manual effort and enhancing efficiency in greenhouse management. These methods enabled sustainable, automated monitoring in large-scale greenhouse environments.

Therefore, solar energy could be a promising source for energy harvesting to mitigate the power demand of smart agriculture systems.

6.2. Wind Energy

A wind turbine is a primary approach to generating electricity from air-flow energy, which can be further used to power various smart agricultural operations. This renewable source of energy reduces dependency on traditional fossil fuels in farming and increases sustainability and productivity.

Wind energy can be harvested using wind belts, which generate electricity via aero-elastic flutter in addition to magnets and coils [114]. An adaptive power-aware routing algorithm (APAR) was proposed to minimize energy consumption during data transfers, thereby extending the network lifespan. The results of data analysis and simulation demonstrated increased efficiency in wind energy harvesting and routing algorithms. The authors also discussed field implementation strategies that suggest farmer collaboration to share resources, making the system more cost-effective. By utilizing wind energy techniques, the sustainability and efficiency of agricultural sensor networks are enhanced.

6.3. Vibration Energy

Vibration energy is a type of mechanical energy generated by vibrations and can be converted into electrical energy. In smart agriculture, small vibrations caused by machinery are captured and used to supply power to sensors or low-energy systems.

A novel energy-harvesting technique that used piezoelectric materials to generate 200 mW of power for agricultural sensor nodes was introduced in [115]. This method improved operation without traditional power sources. The authors also discussed hardware adaptations to optimize the transmission efficiency of data transfer rates with minimal energy use, ensuring that the agricultural system can operate entirely through harvested energy.

6.4. Water Flow Energy

Water flow energy, or hydrokinetic energy, is the free energy harnessed from the movement of water. In smart agriculture systems, electricity can be generated by capturing the kinetic energy from flowing water sources like rivers, streams, canals, or irrigation systems. This renewable energy source promotes sustainability and lowers production costs in farming practices, especially in areas where water is readily available.

A combination of solar, wind, and water flow energy sources utilizing energy harvesting in smart agriculture was introduced in [116]. The power management system charged a 650 mAh NiMH battery to ensure the simultaneous operation of a wireless data acquisition platform (WDAP). The study also explored additional energy storage options like ultra-capacitors and lithium batteries to enhance system longevity.

6.5. Microbial Fuel Cell (MFC) Energy

Microbial fuel cell (MFC) energy utilizes the metabolic processes of microorganisms to generate electricity. MFCs can harness energy from soil, animal manure, or plant waste to provide a renewable energy source for powering low-energy devices such as sensors, controllers, or small monitoring systems. By converting waste into usable energy, MFC technology enhances the sustainability and efficiency of smart farming systems.

Several innovative methods to monitor phreatic aquifers were proposed in [117,125]. A comprehensive system with sub-modules was developed, focusing on performance, energy use, and electric characterization. The system was powered by a terrestrial microbial fuel cell (MFC), making it energy-efficient and sustainable.

7. Differences in Methods for Field and Greenhouse Cultivation

7.1. Field Cultivation

Field cultivation in smart agriculture often relies on mobile platforms such as satellites, airplanes, and UAVs (unmanned aerial vehicles) to gather data on a large scale. These platforms are equipped with various sensors and imaging technologies, including multispectral, hyperspectral, and thermal cameras, to monitor crop health, soil conditions, and pest activity. The flexibility and extensive coverage of these mobile platforms allow for comprehensive monitoring across vast agricultural areas. For instance, IoT-driven precision agriculture systems use land mapping, crop prediction, and irrigation management to optimize resource usage and improve crop yields [126]. Machine-learning algorithms and remote-sensing technologies are integrated to provide real-time data and predictive insights, enhancing decision-making processes in field cultivation.

7.2. Greenhouse Cultivation

In contrast, greenhouse cultivation predominantly uses stationary sensors to monitor environmental conditions such as temperature, humidity, and light levels. These sensors provide continuous data, enabling precise control of the greenhouse microclimate. The controlled environment of greenhouses allows for the implementation of advanced technologies such as automated irrigation and fertilization systems, which ensure optimal growing conditions for plants. Recent advancements in agricultural greenhouse technologies focus on energy management and the integration of IoT devices to enhance productivity and sustainability [127]. Stationary sensors in greenhouses are crucial for maintaining consistent environmental conditions, which are essential for the growth and health of plants.

7.3. Comparison

The primary difference between field and greenhouse cultivation methods lies in the type of sensors and platforms used. Field cultivation benefits from the extensive coverage and flexibility of mobile platforms, while greenhouse cultivation relies on the precision and consistency of stationary sensors. Both methods leverage IoT technologies, AI, and machine learning to optimize agricultural practices, but their applications and implementations vary based on the specific requirements of the cultivation environment.

8. Future Directions

In the face of global challenges, future research in smart agriculture should prioritize advancements in several critical fields to ensure its adaptability and impact. Firstly, integrating artificial intelligence (AI) and machine learning (ML) based on real-time environmental data is essential for creating predictive models that optimize the allocation of water, fertilizers, and other resources. These tools can significantly improve pest and disease-detection systems, making agricultural monitoring more autonomous and precise. AI-driven systems can also help achieve key UN Sustainable Development Goals (SDGs), such as SDG 2 (Zero Hunger) and SDG 13 (Climate Action), by promoting sustainable and efficient resource use.

Secondly, expanding renewable energy sources, including solar, wind, and microbial fuel cells, will be crucial in powering a wider range of agricultural applications like autonomous vehicles and smart irrigation systems. This expansion can help reduce reliance on conventional energy sources, lower costs, and enhance sustainability. Emphasizing these renewable energy solutions supports climate-resilient farming practices, which are essential for adapting to changing environmental conditions.

Thirdly, developing affordable, scalable technologies is vital, especially for applications in developing countries. By creating low-cost sensors, communication networks, and control systems, smart agriculture can offer accessible, real-time decision-making tools to farmers with limited resources. Fourthly, applying wireless communication protocols such as LoRaWAN and ZigBee in agriculture can improve data transmission across extensive and remote farmlands, supporting the scalability of these systems.

Finally, addressing sustainability and climate adaptation is paramount. Future smart agriculture systems must be designed to withstand extreme weather events and unpredictable climate patterns. Innovations in energy-efficient technologies and AI-driven resource management can help farmers conserve water, manage resources during droughts, and respond effectively to climate-related challenges [128].

9. Conclusions

Smart agriculture represents an evolutionary approach to addressing the growing global population’s future resource scarcity and climate change demands. By integrating advanced sensor technologies, Internet of Things (IoT) systems, and renewable energy solutions, smart agriculture can significantly enhance productivity while minimizing environmental impact. These innovations foster sustainable farming practices and promote environmental resilience.

As technological advancements continue incorporating artificial intelligence (AI), machine learning (ML), and scalable designs, smart agriculture will evolve to address complex agricultural needs. AI and ML-driven systems offer promising tools to improve food security by optimizing resource allocation and enhancing pest and disease detection. Additionally, the expansion of renewable energy sources such as solar, wind, and microbial fuel cells supports climate-resilient farming practices, essential for adapting to changing environmental conditions.

Developing affordable and scalable technologies is crucial, especially for applications in developing countries. By creating low-cost sensors, communication networks, and control systems, smart agriculture can provide accessible, real-time decision-making tools to farmers with limited resources. Furthermore, the application of wireless communication protocols like LoRaWAN and ZigBee can improve data transmission across extensive and remote farmlands, supporting the scalability of these systems. This paper has reported innovative approaches towards the development of smart agriculture. The demonstrated technologies, methods, and ideas offer valuable insights for the future development of an efficient smart agriculture industry. These advancements not only enhance productivity and sustainability but also support the global effort to achieve key UN Sustainable Development Goals (SDGs), such as Zero Hunger and Climate Action. By embracing these innovations, smart agriculture can contribute to sustainable growth and environmental resilience worldwide.