Date of Defense
5-6-2026 9:45 AM
Location
E1-1028
Document Type
Thesis Defense
Degree Name
Master of Science in Internet of Things
College
CIT
Department
Computer and Network Engineering
First Advisor
Dr. Farman Ullah
Keywords
Edge AI, Internet of Things (IoT), Smart Irrigation, Precision Agriculture, Deep Learning, LSTM, GRU, RNN, Sustainable Agriculture, Arugula Cultivation, Plant Growth Optimization.
Abstract
Traditional agriculture faces several challenges, including excessive water consumption, inefficient irrigation practices, and the increasing demand for sustainable food production. The rapid advancement of the Internet of Things (IoT) and Artificial Intelligence (AI) provides opportunities to develop intelligent agricultural systems that improve resource management and support precision farming. This thesis presents an Edge AI and IoT-based smart irrigation system designed for real-time environmental monitoring, automated irrigation control, and intelligent irrigation prediction to support the optimal growth of arugula plants. The proposed system continuously monitors key cultivation parameters, including soil moisture, temperature, humidity, light intensity, pH, nutrient levels (NPK), and water consumption, to maintain environmental conditions suitable for healthy arugula growth. The proposed framework integrates multiple sensors connected to an Arduino Uno R4 WiFi board for edge-level data processing and irrigation control. Sensor measurements are continuously collected, monitored, and transmitted to the ThingSpeak cloud platform for real-time visualization, storage, and analysis. The proposed methodology consists of two phases. The first phase focuses on the design and implementation of an IoT-based smart irrigation system that automatically irrigates based on predefined soil moisture thresholds. The second phase employs Deep Learning techniques to analyze time-series sensor data and support intelligent irrigation prediction and decision-making. Two datasets were collected and used in this study. After preprocessing, the datasets were used to train and evaluate Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), and Gated Recurrent Unit (GRU) models using different time-series window configurations. Experimental results demonstrated that the proposed system successfully performs autonomous irrigation, supports continuous environmental monitoring, and enables efficient water management through real-time decision-making while maintaining suitable growing conditions for arugula cultivation. Furthermore, the Deep Learning models effectively captured temporal dependencies in sensor data and accurately predicted irrigation-related parameters. The integration of Edge AI, IoT, cloud computing, and predictive analytics enhances irrigation efficiency, reduces water wastage, improves plant growth conditions, and supports sustainable agricultural practices. The proposed framework advances precision agriculture by providing an intelligent, scalable, and data-driven solution for smart irrigation and plant growth optimization.
Included in
EDGE AI AND IOT-BASED INTELLIGENT SMART IRRIGATION SYSTEM FOR SUSTAINABLE AGRICULTURE
E1-1028
Traditional agriculture faces several challenges, including excessive water consumption, inefficient irrigation practices, and the increasing demand for sustainable food production. The rapid advancement of the Internet of Things (IoT) and Artificial Intelligence (AI) provides opportunities to develop intelligent agricultural systems that improve resource management and support precision farming. This thesis presents an Edge AI and IoT-based smart irrigation system designed for real-time environmental monitoring, automated irrigation control, and intelligent irrigation prediction to support the optimal growth of arugula plants. The proposed system continuously monitors key cultivation parameters, including soil moisture, temperature, humidity, light intensity, pH, nutrient levels (NPK), and water consumption, to maintain environmental conditions suitable for healthy arugula growth. The proposed framework integrates multiple sensors connected to an Arduino Uno R4 WiFi board for edge-level data processing and irrigation control. Sensor measurements are continuously collected, monitored, and transmitted to the ThingSpeak cloud platform for real-time visualization, storage, and analysis. The proposed methodology consists of two phases. The first phase focuses on the design and implementation of an IoT-based smart irrigation system that automatically irrigates based on predefined soil moisture thresholds. The second phase employs Deep Learning techniques to analyze time-series sensor data and support intelligent irrigation prediction and decision-making. Two datasets were collected and used in this study. After preprocessing, the datasets were used to train and evaluate Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), and Gated Recurrent Unit (GRU) models using different time-series window configurations. Experimental results demonstrated that the proposed system successfully performs autonomous irrigation, supports continuous environmental monitoring, and enables efficient water management through real-time decision-making while maintaining suitable growing conditions for arugula cultivation. Furthermore, the Deep Learning models effectively captured temporal dependencies in sensor data and accurately predicted irrigation-related parameters. The integration of Edge AI, IoT, cloud computing, and predictive analytics enhances irrigation efficiency, reduces water wastage, improves plant growth conditions, and supports sustainable agricultural practices. The proposed framework advances precision agriculture by providing an intelligent, scalable, and data-driven solution for smart irrigation and plant growth optimization.