Penulis: Barra Asyqar Rafi; Moechammad Sarosa; Arwin Datumaya Wahyudi SumariArwin Datumaya Wahyudi Sumari

Nama Jurnal: IEEE Xplore

Tahun: 2024

Volume: Tidak ada

Issue: Tidak ada

Halaman: 14-18

Deskripsi: This study explores the development and evaluation of an Internet of Things (IoT)- based automated hydroponic system tailored for lettuce cultivation, aimed at optimizing plant growth while minimizing manual intervention. The system integrates Arduino Nano microcontrollers for managing pH and TDS sensors, transmitting data to a Raspberry Pi 4b for centralized data processing and pump control. Over a rigorous 5-day experimental period, the pH and TDS sensors demonstrated robust performance with average measurement errors of 3% and 4%, respectively, ensuring accurate monitoring of nutrient solution conditions. The automated system effectively maintained optimal pH levels (between 5.5 and 6.5) and TDS concentrations (ranging from 560 to 840 ppm) critical for lettuce growth. Additionally, leveraging the YOLOv5 model for image processing facilitated real-time monitoring and classification of lettuce growth stages, achieving a classification accuracy of 56%. This capability provides insights into plant development and facilitates timely intervention in cultivation practices. These results highlight the potential of IoT and machine learning technologies to enhance precision agriculture by ensuring precise monitoring, efficient resource management, and proactive decision-making in hydroponic farming. By reducing manual labor and optimizing resource utilization, such systems contribute to sustainable agricultural practices, addressing challenges posed by global population growth and urbanization. In conclusion, this research underscores the feasibility and benefits of integrating IoT and AI technologies in modern agriculture, paving the way for future advancements in automated farming systems that can sustainably meet global food demands.

Pendahuluan: The rapid growth of the global population, projected to reach 9.7 billion by 2050, along with increasing urbanization, has placed significant pressure on the world's food systems [1]. Urbanization not only causes an increase in population numbers but also results in limited space for conventional agriculture, especially in urban areas [2]. The challenge of providing sufficient, high-quality, and sustainable food requires innovative agricultural approaches. One increasingly popular solution is hydroponics, a method of growing plants without soil using a nutrient-rich solution. Hydroponic systems have several advantages, such as allowing plants to grow faster and produce higher yields compared to conventional methods, while also requiring less land and being applicable in controlled environments, making them ideal for densely populated urban areas [3][4]. Lettuce (Lactuca sativa L.), a member of the Asteraceae family, is one of the most suitable plants for hydroponic cultivation due to its fast growth cycle and relatively small space requirements [5][6]. Additionally, lettuce is rich in essential nutrients such as minerals and organic compounds, making it a popular choice among consumers seeking healthy food options [7][8]. However, despite the advantages of hydroponic cultivation, there are still major challenges in maintaining optimal conditions for plant growth. Critical parameters such as acidity (pH) and nutrient concentration in the solution (Total Dissolved Solids or TDS) must be strictly monitored and controlled to ensure healthy plant growth. For optimal lettuce growth, the solution's pH must be maintained within a range of 5.5 to 6.5, while TDS should be between 560 and 840 ppm [9][10]. If the solution exceeds these thresholds, it can result in root damage, leaf burn, and plant death. Visually, healthy roots will appear clean white and spread in all directions. Symptoms of nutrient overdose include yellowing roots that all hang down. Consequently, leaves begin to wilt, their color starts to fade, and in a short time, the plant will die. To address these challenges, Internet of Things (IoT) technology has emerged as a promising solution with the ability to automate real-time monitoring and control of environmental conditions [11]. IoT enables the integration of various sensors that can continuously monitor parameters such as pH, TDS, and temperature and transmit this data to a control center for further analysis and corrective action [12]. In this context, the use of microcontrollers like the Arduino Nano to operate sensors and single-board computers like the Raspberry Pi 4b for remote data management and monitoring is highly relevant. This combination of devices enables accurate data collection and precise control of the hydroponic environment, ultimately enhancing the efficiency and yield of plant production. This study offers a novel approach by integrating IoT technology with the YOLOv5 image recognition algorithm for real-time monitoring of lettuce plant growth. The YOLOv5 algorithm is chosen for its high speed, efficiency, and ability to detect and classify lettuce growth stages with adequate accuracy, even in limited hardware conditions like the Raspberry Pi 4b [13][14]. These advantages allow the system to identify and respond to changes in plant growth stages promptly, which is crucial in a precision- controlled hydroponic environment. The use of YOLOv5 in this hydroponic system not only provides real-time plant growth data but also enables automated maintenance decisions, such as adjusting nutrient and pH levels based on observed plant conditions. Furthermore, with the increasing demand for lowmaintenance agricultural systems in urban environments, this research aims to develop and implement an IoT-based hydroponic lettuce system that is not only efficient but also easy to manage. The system is designed to minimize human intervention by automating the monitoring and management of the plant growth environment, which is highly beneficial for farmers and hydroponic garden managers operating in urban areas with limited time and space. Additionally, this research seeks to fill gaps in the existing literature by exploring the integration of IoT technology and machine learning in hydroponic agriculture. While many studies have highlighted the benefits of IoT in agriculture, few have discussed its integration with image recognition algorithms like YOLOv5 for real-time plant growth monitoring [15][16]. The YOLOv5 algorithm is chosen for its high speed and efficiency, as well as its ability to detect and classify the maturity levels of lettuce plants in real-time [17]. YOLOv5 also has high accuracy and can run on limited hardware like the Raspberry Pi 4b [18], making it ideal for this application. Thus, this research not only contributes to improving the efficiency and yield of hydroponics but also provides a foundation for further development in the use of advanced technology in modern agriculture. This research is expected to significantly contribute to the application of IoT technology in modern agriculture and pave the way for further developments in this field. The results obtained from this research could serve as a basis for implementing similar systems on a larger scale, which would not only enhance agricultural productivity but also reduce the environmental impact caused by conventional farming methods.

Kata Kunci: YOLO; Accuracy; Hydroponics; Sensor systems; Real-time systems; Sensors; Resource management; Monitoring; Farming;

Total Kunjungan: 44 kali