EE/CE Capstone Projects

Abstract: The ASCS prototype is comprised of a battery-powered sensor device for recording EEG (electrical brain activity) signals and a machine learning algorithm for classifying and organizing gathered data in a research-friendly format. The ASCS is a low cost alternative to similar EEG recording devices with greater potential for accuracy than non-EEG-based sleep classifying devices like FitBit or Apple Watch. 

Abstract: The goal of this project is to create a charging system for 12V batteries on a sailboat. Currently, there exist systems to do this based on rotational power from a combustion engine, systems based on gathering energy from solar power, and high-power wind systems. We aim to move away from using fuel systems, solar panels that may not operate as frequently as we need, and from the high-cost, high-precision market systems for wind generation. Because there is not always enough wind for high power charging, our system will operate in a broader range of conditions to consistently supply a small trickle charge to the battery bank. The system will take power in from a small-scale Wind Turbine (WT). This power will then be converted to charge the shipboard battery. The system will also have controls to monitor and control the power flow and stop the battery from overcharging. The control system will also display the monitored data on a screen. The focus of this project will be using a variable power source (the WT) to charge a battery. The intention is to build the WT as well but is of secondary importance to the charge controller.

Abstract: Hydro-generators, known for their sustainable, reliable energy production, can be optimized through advanced control and conversion methods. This capstone project integrates a DC-DC converter with pulse-width modulation (PWM) within a hydro-generator system, aimed at boosting energy conversion efficiency and aligned with battery charging requirements. This DC-DC buck converter takes variable hydro-generator voltage as an input and uses current control to meet the requirements of a lead acid 12V battery. PWM control is used to govern the input current to the storage system, ensuring smooth charging of the battery as well as minimizing harmonic distortions in the voltage and current waveforms. By adopting this technology, hydro-generators can significantly bolster their contribution to the efficiency and stability of renewable energy systems. Not only is this project applicable to households in rural areas needing renewable energy, but it is also ready to scale up for use with complex smart grid power conversions in the future.

Abstract: The project is a solar battery charging circuit that uses a 50W solar panel to charge a 12v battery. The charging is achieved using a buck converter circuit controlled by an Arduino Uno. The battery is then used to periodically power a thermal imaging camera that detects hotspots resulting from soiling on the solar panel. Hotspots are a common issue on solar panels and cause degradation of the panels over time. Hotspot detection is critical for increasing the lifespan of a solar panel.

Abstract: ​​​​​​A project to develop a scalable, Arduino-controlled, off-grid solar power system aimed at charging electronic devices, particularly in rural areas of developing countries and regions prone to power outages. The system, designed to improve electricity access, consists of two main modules:

1. Solar Charging Module: Features two 50-Watt solar panels and an Arduino Mega to charge four 18650 lithium-ion batteries. It includes an SD card for logging data from various sensors and a buck converter for efficient power management. The module uses PWM control via a P&O algorithm for optimal charging.
2. Device Charging Module: Utilizes another set of four 18650 lithium-ion batteries, controlled by an Arduino Uno, to regulate a USB-C port that converts 14 Volts to 5 Volts, suitable for charging devices such as cell phones.

This system is particularly beneficial for individuals who have mobile connectivity but face challenges in accessing reliable electricity.

Abstract: Focuses on the development of a State of Charge (SoC) Algorithm Controller for an EEG-Based Sleep Tracker Battery. The primary objective is to design and implement a charging device that utilizes a SoC Estimation Algorithm to optimize the charging process of the battery. The project will be conducted in multiple stages, including simulation using Simulink and breadboard testing, to determine the optimal RLC components and PWM frequency and duty cycles for the buck converter. In the initial stages, the project will concentrate on the design and simulation of the charging system. Simulink will be employed to model and analyze the performance of the SoC Algorithm Controller under various conditions. Breadboard testing will be conducted to validate the simulation results and fine-tune the system parameters. The selection of appropriate RLC components and the optimization of PWM frequency and duty cycles will be crucial in ensuring the efficiency and reliability of the buck converter. As the project progresses, emphasis will be placed on the form factor and the design of the Perf Board for the charger. Research will be conducted to explore different methods for implementing SoC estimation algorithms, which will be used to control the charging rate of the battery. To achieve precise control over the charging process, a current-controlled device will be integrated with the microcontroller. The successful completion of this project will result in the development of an efficient and reliable SoC Algorithm Controller for the EEG-Based Sleep Tracker Battery.

Abstract: Our project is to design an AI-driven vehicle that will follow a user, allowing the user to store items in an area on the top, essentially creating a cart-like experience. To track the user's location, the user will hold a phone that transmits a Bluetooth signal to the cart. The AI will use the signal strength to determine the user's proximity. Additionally, ultrasonic sensors in front and behind the cart will provide information to the AI, helping it detect obstacles for safety. As for powering the cart a battery bank composed of lithium-ion batteries will be used. The power is regulated through a custom buck converter which lowers the voltage coming from the batteries. The buck converter is controlled using an Arduino that calculates how much power the batteries have left and maintains a constant voltage. That converted power then enters a motor driver which powers the cart’s motors.

Abstract: The Automatic Pet Feeder (APF) is a pet feeding solution which is designed to aid busy pet owners. The APF allows pet owners to remotely schedule and manage their pet’s meals right from their web browser. It is designed to accommodate various types of pets, allowing users to choose the time, frequency, and portion size of each meal. The focus is to offer convenience for pet owners when they are unable to be present to feed their animals. It provides reassurance that their pets are receiving accurate and timely meals. 

Abstract: The Bluetooth Signal Combiner aims to create a hub device that merges audio from multiple Bluetooth-connected devices (e.g., laptops, smartphones) and transmits it to one or more sets of Bluetooth headphones. Targeting multitasking students or workers, this solution allows simultaneous audio playback from various sources through the same pair of headphones. Unlike existing Bluetooth multipoint technology, this hub facilitates a seamless connection and mixing of Bluetooth Classic audio devices. 

Abstract: The Beamforming Microphone Array (BFA) leverages multiple microphones to selectively listen to sounds from a specific direction. By controlling the multiple signals' phase and amplitude differences, the BFA can effectively steer its own sensitivity pattern creating a beam towards a specific direction. LEDs from around the BFA indicate the direction where the beam is pointed towards, and a knob at the center of the BFA allows the user to adjust the beam location and volume. 

Abstract: The SAGE System is a Semi-Autonomous Growing Environment tailored for easy home cultivation of plants, ideal for those without access to outdoor gardening spaces. Equipped with internal sensors, it monitors air temperature, humidity, soil temperature, and moisture, allowing users to set optimal conditions. It includes a lighting schedule managed by two full-spectrum LED light modules adaptable for different plant growth stages, and features a water reservoir with an LED indicator to alert users when refilling is needed, though it mainly waters automatically based on soil moisture levels.

This system aims to simplify home cultivation of edible plants and fungi for individuals with limited space, like college students or office workers in urban areas. It also acts as a proof of concept for a fully automated growing chamber potentially suitable for interplanetary travel, leveraging recent advances in CAD and 3D printing technology for space exploration applications.

Abstract: The H.O.S.S (Haptics Object Sensor System) is a device intended to detect objects at varying distance thresholds, with Eccentric Rotating Mass (ERM) motors to provide haptic responses for each of the thresholds. The HOSS features a repurposed millimeter wave object detection module used to estimate the Time of Arrival (TOA) of the reflected waves detected from objects. The HOSS is powered by coin cell batteries so that it can be replaced by the user. An accelerometer is included to ensure that the device is parallel to the ground when reading for reflected waves.

Abstract: The Automated Greenhouse is an enclosure that provides its user a stable environment for household plants to thrive by providing adequate soil moisture, temperature, and humidity levels. The system will allow the user to set defined levels for these parameters, which will be maintained through consistent monitoring and corrective action if levels drift from the defined range. It also provides a daily lighting schedule to ensure adequate light is provided.

Abstract: The modular mechanical keyboard seeks to suit a wide range of typing, layout, functional, and aesthetic preferences, so as to be configurable and customizable to most user’s needs. The modularity primarily comes from the side and top mounting modules, which extend the keyboard’s functionality. These modules can range from a numpad, to navigation keys, and even a function row. Other features include multi-region layout support, Bluetooth connectivity, multiple USB-C ports, “under-glow” lightning, a display, and much more. 

Abstract: This project is the design and construction of a non-centralized matrix of soil probes that gather information from various sensors on soil moisture level, sun exposure, and temperature. This matrix is designed to provide a fast and accurate response that allows the user(s) to make the best decision for the care and growth of differing plant life in either an indoor or outdoor environment. 

Abstract: The Smart Monitoring of Air Quality (SMAQ) device serves as a proof of concept indoor air quality monitoring system to keep users up to date on metrics such as temperature, humidity, carbon dioxide, and particle matter in their homes via a web implementation. The system combines hardware design, networking, and machine learning to calculate the air quality index indoors, and if classified as harmful, works to improve conditions by turning on a correction unit such as a fan to bring metrics to healthy levels. 

Abstract: The Sign Language Portable Interpreter (SLaPI) is a compact interpreter for American sign language. SLaPI uses computer vision and artificial intelligence algorithms to translate gestures into English text, streamlining communication between the deaf and the hearing. The device is compact and lightweight so that it is easy to pick up and move to wherever an interpreter is needed.

Abstract: The Vehicle Cooperative Perception System (VCPS) project aims to improve road safety through advanced environmental interpretation and reliable vehicle communication. Integrated into PACCAR's trucks, the system uses lidar and camera sensors to monitor surroundings and share vehicle and pedestrian locations. Utilizing state-of-the-art neural networks and other AI techniques, it outperforms traditional computer vision methods. This collaborative perception allows proactive hazard identification and wireless notifications between nearby vehicles, which in the future can be used for safe decision-making. Implementing edge computing technology directly in the vehicle enables real-time object detection and localization from lidar, cameras, and GPS. The project also explores V2X communication technologies like C-V2X (LTE & 5G), DSRC, and Wi-Fi. Designed with energy efficiency in mind, the system supports the transition to electric vehicles while advancing the future of intelligent driving. 

Abstract: The Disc Golf Guardian is a smart disc golf basket which implements accelerometers to track chain impact locations. The DGG will come equipped with three putting games that will encourage players to hone their putting abilities while having fun. DGG will interface with users through 4 buttons and a LED dot matrix display. 

Abstract: WELS monitors temperature, humidity, and light level and is wirelessly connected to a building’s control system though a IEEE 15.4 network.  WELS offers extended life compared to other battery powered fully wireless options on the market due to light and radio frequency energy harvesting. Fitting into existing control systems, WELS connects to a building’s 915MHz IoT network, and sends sensor readings to a server. While an IoT sensor is nothing new, the novel aspect of WELS is radio frequency energy harvesting in the 2.45GHz ISM band. Ambient RF signals will be captured by an integrated antenna, and used to extend the life of the device past what the included battery alone could provide. Ideally, WELS will be able to operate for multiple years without service. 

Abstract: The goal of this project is to create a charging system for 12V batteries on a sailboat. Currently, there exist systems to do this based on rotational power from a combustion engine, systems based on gathering energy from solar power, and high-power wind systems. We aim to move away from using fuel systems, solar panels that may not operate as frequently as we need, and from the high-cost, high-precision market systems for wind generation. Because there is not always enough wind for high power charging, our system will operate in a broader range of conditions to consistently supply a small trickle charge to the battery bank. The system will take power in from a small-scale Wind Turbine (WT). This power will then be converted to charge the shipboard battery. The system will also have controls to monitor and control the power flow and stop the battery from overcharging. The control system will also display the monitored data on a screen. The focus of this project will be using a variable power source (the WT) to charge a battery. The intention is to build the WT as well but is of secondary importance to the charge controller.

While backpack hiking, it is very important to have a reliable way to obtain drinkable water. Even for shorter hikes, carrying enough water for a day is very heavy and not very practical. There are currently a large variety of products to solve this problem, with a large range of prices and reliability. The project's objective is to build a fully functional water purifier system that balances price and reliability, while also being solely powered by a small set of solar panels that can be easily attached to the user's backpack.