Find a project you can actually build
Working IoT, AI/ML and embedded builds with abstracts, component lists, documentation, and developer support — ready for your final-year submission.
piezo electric shoes
Traditional methods for measuring cholesterol and hemoglobin levels are invasive, requiring blood samples and laboratory processing, which may be inconvenient for continuous or remote health monitoring. This project proposes a non-invasive health monitoring system that estimates cholesterol and hemoglobin levels using Photoplethysmography (PPG) signals, combined with machine learning (ML) algorithms. The system is built using a NodeMCU microcontroller and integrates an SpO₂ sensor and heart rate sensor to capture real-time PPG waveforms. These signals are analyzed using a Flask-based server, where ML models trained on clinical datasets predict the user’s cholesterol and hemoglobin levels with reasonable accuracy. The results are displayed on a secure web dashboard for remote health monitoring. This contactless solution supports continuous assessment without needles or lab intervention, making it ideal for home use, remote clinics, and chronic care settings. It provides an accessible, low-cost alternative for routine health diagnostics while promoting user comfort and reducing healthcare burdens.
blind moss code conversion wrist band
Visually impaired individuals often face challenges in real-time communication, especially in noisy or inaccessible environments. To address this, the proposed system introduces a Blind Morse Code Conversion Wristband that enables intuitive and silent communication using Morse code, vibration feedback, and machine learning. Built around an ESP32 microcontroller, the system includes a mic module to capture short and long sound pulses (dots and dashes), which are interpreted using machine learning algorithms hosted on a Flask-based server. Recognized Morse code is then converted into text or pre-programmed messages and delivered to the user through a vibration motor, allowing discreet message reception. Additionally, a speaker (SPEA) provides optional audio output for accessibility and feedback. The wristband operates as a two-way assistive communication tool, allowing visually impaired users to send or receive messages silently. The integration of IoT and ML ensures real-time processing, adaptability to user patterns, and remote monitoring capabilities. This system represents a low-cost, wearable communication aid designed to enhance independence and accessibility for the blind and visually impaired.
Smart Infant Incubator with Automated Oxygen and Temperature Control and Real-Time Alerts
Premature and critically ill infants require precise environmental conditions for survival and recovery. Traditional incubators often lack remote monitoring and intelligent control features, increasing the risk of manual error and delayed intervention. This project presents a Smart Infant Incubator System using IoT and Flask, designed for automated control of oxygen and temperature levels with real-time alerts to caregivers. The system is built around an ESP32 microcontroller, which continuously monitors vital parameters using a SpO₂ sensor to track blood oxygen saturation and a temperature sensor to ensure thermal stability inside the incubator. Based on real-time sensor data, a solenoid valve regulates oxygen flow, while a water pump is used for humidification or temperature adjustment mechanisms within the chamber. All data is transmitted to a Flask-based IoT dashboard, enabling remote visualization, threshold-based alerts, and historical trend analysis. This intelligent incubator improves neonatal care by reducing the need for constant manual supervision and enhancing response time during critical fluctuations in an infant’s condition.
accident prevention hairpin bend
Hairpin bends are among the most accident-prone areas on hilly roads due to poor visibility, sharp turns, and high-speed vehicle movement. To address this challenge, this project proposes an IoT and Machine Learning-based Accident Prevention System for Hairpin Bends that provides early warnings and real-time alerts to prevent collisions. The system uses a NodeMCU and ESP32-CAM module to detect incoming vehicles at critical turning points. The ESP32-CAM captures live video feeds, which are processed via a Flask-based server using machine learning algorithms to detect vehicles and their direction of movement. A vibration sensor monitors ground impact to detect sudden vehicle braking or crashes. When a vehicle is detected approaching from the opposite side, the system activates a traffic light, LCD (16x2) warning display, and buzzer to alert drivers in real time. The system also includes GSM SIM 800 and GPS modules for sending the location and incident alerts to nearby authorities or emergency services. Wired connections and a secured hardware setup ensure reliable performance in rough terrains. This solution offers a cost-effective, automated, and scalable method to enhance road safety in mountainous and blind-spot regions.
Non invasive Measurement of cholesterol and haemoglobin using ppg
Traditional methods for measuring cholesterol and hemoglobin levels are invasive, requiring blood samples and laboratory processing, which may be inconvenient for continuous or remote health monitoring. This project proposes a non-invasive health monitoring system that estimates cholesterol and hemoglobin levels using Photoplethysmography (PPG) signals, combined with machine learning (ML) algorithms. The system is built using a NodeMCU microcontroller and integrates an SpO₂ sensor and heart rate sensor to capture real-time PPG waveforms. These signals are analyzed using a Flask-based server, where ML models trained on clinical datasets predict the user’s cholesterol and hemoglobin levels with reasonable accuracy. The results are displayed on a secure web dashboard for remote health monitoring. This contactless solution supports continuous assessment without needles or lab intervention, making it ideal for home use, remote clinics, and chronic care settings. It provides an accessible, low-cost alternative for routine health diagnostics while promoting user comfort and reducing healthcare burdens.
Farm wildlife detection and agriculture
Wildlife intrusion into agricultural fields poses a significant threat to crop health and farmer safety, while poor environmental monitoring can lead to inefficient irrigation and yield loss. This project presents an IoT and Machine Learning-based Smart Agriculture System that integrates wildlife detection with agro-environmental monitoring and automation. At its core, the system uses an ESP32 microcontroller to interface with a suite of sensors and actuators. An ESP32-CAM module captures real-time visuals of the farm, where machine learning algorithms detect the presence of animals. Upon detection, a servo motor activates a deterrent mechanism, and a buzzer alerts nearby workers. Additionally, the system monitors soil and crop conditions using a soil moisture sensor, temperature sensor, and color sensor, ensuring optimal growth conditions. A water pump connected to a hose delivers irrigation automatically based on soil data. The GPS module provides location tracking, while the GSM 800 module sends SMS alerts to the farmer in case of wildlife intrusion or critical environmental thresholds. Data is logged and visualized on a Flask-based IoT dashboard, enabling remote monitoring and control. This integrated solution promotes smart farming by enhancing crop safety, automating irrigation, and enabling real-time environmental and wildlife awareness.
Automated Trash Compactor with Alert System
Overflowing public trash bins are a major concern in urban waste management, often leading to environmental hazards and operational inefficiencies. This project presents an Automated Trash Compactor with an Alert System, designed to optimize waste volume reduction and enable real-time monitoring using IoT and Flask technologies. The system is built around the ESP32 microcontroller, which collects data from an ultrasonic sensor mounted inside a smart dustbin (Dustbin SQ) to measure fill levels continuously. When the trash reaches a predefined threshold, a high-power DC motor is activated to drive a pully and rope mechanism, compressing the waste within the bin. A dustbin cover ensures safety during compaction. The system also sends real-time status updates and alerts to municipal authorities or waste handlers via a Flask-based IoT dashboard, ensuring timely disposal and maintenance. This automation reduces the need for frequent manual checks, increases bin capacity through compaction, and supports cleaner and more efficient waste management in smart city environments.
Helmet-Integrated Motorcycle Safety System with Impact Detection and Anti-Reflective Visor
Motorcycle riders are highly vulnerable to road accidents, and the lack of integrated safety features in helmets contributes significantly to the severity of injuries. This project presents an IoT-enabled Helmet-Integrated Motorcycle Safety System that enhances rider protection through impact detection, alcohol level sensing, and vision improvement using smart visor technologies. The system is built using a NodeMCU microcontroller for real-time data acquisition and remote communication. A copper coil embedded in the helmet structure detects head impacts through induced signals, triggering emergency protocols. An alcohol sensor monitors the rider’s breath for signs of intoxication, and a relay module ensures ignition control based on safety conditions. A GSM SIM 900A module sends automatic SMS alerts to emergency contacts with live GPS coordinates in the event of a crash or unsafe alcohol detection. Additionally, the helmet is equipped with anti-reflective film and water-repellent coating on the visor to improve visibility during harsh weather or night rides. All data and alerts are managed through a Flask-based IoT dashboard, enabling remote monitoring and ensuring compliance with safety standards. The system offers a low-cost, scalable solution aimed at significantly reducing fatalities and enhancing rider awareness and safety.
Classifying animal behavior from accelerometry data via recurrent neural networks
Understanding and monitoring animal behavior is essential in wildlife research, livestock management, and health diagnostics. Traditional observation methods are labor-intensive and limited in scope. This project proposes an IoT-based intelligent animal behavior classification system using accelerometry data processed through Recurrent Neural Networks (RNNs). The system is powered by an ESP32 microcontroller and incorporates a 3-axis accelerometer to capture movement patterns, a temperature sensor, a heart rate sensor to monitor physiological parameters, and a hall effect sensor to detect movement or rotational activity. A battery monitoring circuit ensures continuous system availability during remote deployment. The collected data is transmitted to a Flask-based IoT server, where it is preprocessed and classified using a trained RNN model capable of identifying different behavioral states such as resting, walking, running, or abnormal activities. Real-time analytics and visualization are made accessible through a web dashboard, supporting remote monitoring and long-term behavior tracking. This system demonstrates a robust, scalable, and autonomous approach for animal behavior classification using AI and sensor fusion, suitable for wildlife conservation, smart farming, and veterinary health monitoring.
ship sink analysis
Maritime safety remains a critical concern as ships often face unpredictable environmental conditions that can lead to disasters such as sinking. Traditional ship monitoring systems lack the responsiveness and intelligence needed to detect early signs of structural failure or critical environmental changes. This project presents an IoT-based Ship Sink Analysis System that enables real-time monitoring and predictive analysis of sinking conditions using a variety of onboard sensors. The system is built around the ESP32 microcontroller, which collects data from a temperature sensor, air pressure sensor, water pressure sensor, and a 3-axis accelerometer to detect abnormal tilting or impact. A GPS module is used to provide the ship's real-time location, especially useful in emergencies. The setup is simulated using a container with a water pump to replicate varying water levels and pressure conditions. All sensor data is transmitted to a Flask-based IoT dashboard, enabling remote monitoring, data logging, and early warning alerts. This solution supports early detection of potential hazards, improves crew response time, and enhances overall maritime safety by enabling predictive sink analysis.
smart gas stove
Gas stoves are widely used in households and commercial kitchens, yet they pose significant safety risks due to gas leaks, unattended cooking, and inefficient ignition. This project proposes a Smart Gas Stove System that integrates IoT and automation to enhance cooking safety, efficiency, and remote monitoring. The system is built around an ESP32 microcontroller, which controls the stove's ignition and monitors gas leakage in real-time. A gas sensor continuously checks for hazardous leaks, and upon detection, the system can trigger alerts and shut off the supply automatically. A servo motor is used to automate the ignition mechanism in coordination with an electronic lighter, allowing the stove to ignite safely and efficiently. A 16×2 LCD display provides real-time feedback on system status, including gas concentration levels and operation modes. The entire system is mounted on a custom-designed PCB board for compact integration and reliability. Additionally, a Flask-based IoT dashboard enables remote control and monitoring, allowing users to supervise the stove's activity from a web interface. This intelligent gas stove solution prioritizes user safety, energy efficiency, and modern kitchen automation.