Enhancing Fall Detection Accuracy in Diverse Home Environments using an Adafruit Microcontroller-Based Wearable Device
James Toche, Tyler Boulom · 2024
The problem addressed in this project is the variability in fall detection accuracy due to different room layouts in homes. Traditional fall detection systems often fail to account for diverse environments, leading to false positives or missed detections. To solve this, the project integrates an Adafruit microcontroller with an accelerometer to monitor movement and detect falls, using a threshold-based algorithm for accuracy. The key components include the microcontroller, battery, and casing, which allow for wearable, real-time monitoring. Challenges included ensuring consistent performance across various room setups. These were addressed through controlled experiments simulating different living room and kitchen layouts. Results demonstrated that the device could accurately detect falls in various environments, with minimal false positives. This solution offers a reliable and adaptable fall detection system, making it highly useful for elderly individuals or those at risk of falls, ultimately enhancing safety in diverse home environments.