Project
| # | Title | Team Members | TA | Documents | Sponsor |
|---|---|---|---|---|---|
| 8 | Backpack Buddy - Wearable Proximity/Incident Detection for Nighttime Safety |
Emily Grob Jeric Cuasay Rahul Kajjam |
Zicheng Ma | design_document1.pdf final_paper1.pdf photo1.PNG photo2.png presentation1.pdf proposal1.pdf video |
|
| # Backpack Buddy Team Members: - Student 1 (cuasay2) - Student 2 (rkajjam2) - Student 3 (eegrob2) # Problem The UIUC campus is relatively a safe place. We have emergency buttons throughout campus and security personnel available regularly. However, crime still occurs and affects students walking alone, especially at night. Staying up late at night working in a classroom or other building can lead to a long scary walk home. Especially when the weather is colder, the streets are generally less populated and walking home at night can feel more dangerous due to the isolation. # Solution A wearable system that uses night vision camera sensor and machine learning/intelligence image processing techniques to detect pedestrians approaching the user at an abnormal speed or angle that may be out of sight. The system would vibrate to alert them to look around and check their surroundings. # Solution Components ## Subsystem 1 - Processing Processing Broadcom BCM2711 SoC with a 64-bit quad-core ARM Cortex-A72 processor or potentially an internal microprocessor such as the LPC15xx series for image processing and voltage step-down to various sensors and actuators ## Subsystem 2 - Power Power Converts external battery power to required voltage demands of on-system chips ## Subsystem 3 - Sensors Sensors Camera - Night Vision Camera Adjustable-Focus Module 5MP OV5647 to detect objects in the dark Proximity sensor - detects obstacle distance before turning camera on, potentially ultrasonic or passive infrared sensors such as the HC-SR04 Haptic feedback - Vibrating Mini Motor Disc [ADA1201] to alert user something was identified # Criterion For Success The Backpack Buddy will provide an image based solution for identifying any imposing figure within the user's blind spots to help ensure the safety of our user. Our solution is unique as there currently no wearable visual monitoring solutions for night-time safety. potential stuff: Potentially: GNSS for location tracking, light sensor for outdoors identification, and heartbeat for user stress levels camera stabilization heat camera |
|||||