Capstone Project
| Group | 2023-18 | Status | inprogress |
| Title | Advanced Hydroponics Control and Monitoring System | ||
| Supervisor | R. Coutinho | ||
| Description | An advanced hydroponics system using the NFT (Nutrient Film
Technique) model is to be developed. At its core, this project revolves around ensuring optimal plant
growth by leveraging a series of sensors such as pH, EC, TDS, thermometers, and
humidity detectors. These sensors will provide real-time data about the
environment, enabling our system to adjust conditions automatically. Using a
custom designed Printed Circuit Board (PCB), actuators like
humidifiers, pumps, and smart valves will be integrated. Control will be taken to the next level by
adding grow lights and even cameras to monitor plant health visually.
A web platform will be created, along with a cross-platform mobile app, developed in React, Node.js, CSS, HTML, and React Native. These platforms will connect with our hardware components via Firebase for remote access. Our platform will provide analysis tools. Growers will be able to look at stored samples, presented on easily understandable graphs, and even have access to past growth cycles. This aids in fine-tuning the environment for future growth cycles. Another core feature is the integration of machine learning. By taking daily snapshots of the grow space with a camera, these images will be analyzed to spot deficiencies or disease patterns on the plants. Using Convolutional Neural Networks (CNNs) as a model and integrating AWS services to process these images, the systems aims to detect and alert growers to any potential issues. |
||
| Requirement | ● Programming Languages: ○ Python: ■ MicroPython for the PCB programming ■ AWS services like Lambda or SageMaker. ○ JavaScript: ■ React.js ■ Node.js ○ HTML/CSS ○ React Native ● Cloud Platforms: ○ Firebase: ■ Real-time database ■ Firestore for data storage ■ Firebase Storage. ○ AWS: ■ SageMaker ■ Lambda ● Machine Learning: ○ Developing Convolutional Neural Networks (CNNs) model. ○ Training, and deploying ML model using AWS SageMaker. ● Image Processing: ○ Image and color recognition techniques ● User Interface and User Experience (UI/UX) Design: ○ User-friendly design for website and mobile platform | ||
| Tools | ● Hardware: ○ Sensors (pH, EC, TDS, DHT22,...) ○ Actuators (ultrasonic humidifier, cameras, grow light, peristaltic pumps, …) ○ Custom-built PCBs with USB to serial communication. ○ Hydroponics equipment for enclosure and grow medium ● Software: ○ Thonny: ■ MicroPython for PCB programming ○ PyCharm, Webstorm: ■ React, Node.js, CSS, HTML for website ■ React Native for mobile app ■ Python for machine learning ○ Firebase: ■ Real-time database ■ Firestore ■ Storage ○ AWS: (can be replaced if easier method found) ■ Lambda for serverless functions. ■ SageMaker for custom machine learning models. ■ Rekognition for image analysis. | ||
| Number of Students | 5-6 | ||
| Students | B. Bélancourt, A.H.W. Law, P. Riachi, Z. W. Wang, T. Thiyagaraja Iyer, T. Blogu | ||
| Comments: | |||
| Links: | |||