Course Overview

This course focuses on scientific research and engineering. During the first part of the year, students conduct independent research projects that incorporate reviewing literature, making conjectures, developing methodology, designing experiments, and communicating findings. Their final projects are presented at a school-wide science fair, with the possibility for advancement to regional, state, and international fairs. During the second part of the year, students work in small teams in order to engineer new products – usually assistive technology devices. They meet with clients, conduct patent searches, design and build prototypes, demonstrate their products to expert judges, and deliver the products to their clients. Throughout the course, students practice incorporating purpose, clarity, organization, mechanics, and audience appeal as they communicate about topics in science and technology. Assignments consist of research papers, short essays, technical reports, and presentations. Students participate actively, as both writers and self-editors, and their works are consistently revised and often submitted for publication in online and print journals.


Meet The Cognitive Drivers!

The Cognitive Drivers!

CMO Suhruth Vuppala (top left), CIO Krisha Patel (bottom left), CTO Anne Tie (middle), and CEO Anyee Li (right). Additionally, Dr. Kevin Crowthers was our advisor.


Problem Statement: Operating a motor vehicle while sleepy, also known as drowsy driving, is a prevalent problem in the United States. Unfortunately, drowsy driving is a major contributor to a large number of vehicle accidents. Drowsy drivers often experience similar symptoms to those who drive under the influence of alcohol; their reaction time and awareness decreases as the driver becomes drowsier.

Motivation: We aim to decrease vehicle accidents on the road by preventing drivers from falling asleep at the wheel. Our device is directed towards all vehicle owners, especially individuals who have diagnosed sleeping conditions and are more prone to sleeping on the wheel.


Design Approach

During March and April, we dived into background/market research to understand the key features we wanted in our product. From there, we created three preliminary designs. In the end, we decided on doing a dual detection system which was design 1 as it met our most crucial requirements.

Design 1


Build Process

(1) Detection mechanisms were finalized and initial design sketches of detection systems with mounting apparatus were generated. (2) Two mounting apparatuses, a vent clip and rear-view mirror mount were designed in CAD model for the function of holding the web-camera system in the car. (3) After testing for functionality in the car, the rear-view mirror mount was finalized as official mounting mechanism. (4) Two webcams were attached to camera extension of rear-view mirror mount. (5) Eye detection and lane detection algorithms were implemented and tested for functionality when computer was hooked up to webcam. (6) Once thresholds and code were finalized after series of tests, code was tested on the Raspberry Pi. (7) Full device with Raspberry Pi was inserted into vehicle for testing. (8) Verbal feedback from client was taken and put into redevelopment of future prototypes. (9) Iterated with a new prototype with revised algorithms, improved user-feedback loop, and alert mechanisms.

Lane Detection Eye Detection Assembled Camera Mount


Final Device

The final design prototype involves a 3D printed rearview mirror mount. Attached to the mirror mount is a camera extension piece to which two web USB cameras are attached using hot glue. These cameras are arranged in such a way that the front camera angle can be adjusted to work well for different cars, and the back camera is stoic to ensure its view remains on the road. These cameras are then connected to a Raspberry Pi microcontroller that runs two computer vision programs. The first, which connects to the front camera, detects the openness of a driver's eyes, the second, which connects to the back camera, detects whether a driver is remaining in their lane. If the driver’s eyes are not detected to be open for a certain period of time, or if their car is detected to not be in their lane correctly, the device will be prompted to sound an alert. This alert is a sound played through a speaker connected to the Raspberry Pi through bluetooth. This speaker is placed on the dashboard. The Pi is powered from a car through a 5V port when a car is turned on. The Pi loses power when the car is shut off.

Final Device



For more information on care/maintenance or how to replicate our device, check out the document below. For any further questions, feel free to contact me (, Anyee (, Krisha (, or Suhruth (


Final Presentation

Below is the Acceptance and Delivery Review presentation.