Assessing the Feasibility of Augmenting Fall Detection Systems by Relying on UWB-Based Position Tracking and a Home...

Area of Science:

• Gerontology
• Robotics
• Biomedical Engineering
• Computer Science

Background:

• Falls are a major cause of injury in older adults, with current detection systems often generating false alarms.
• High sensitivity in fall detection systems leads to numerous false alarms, straining emergency resources and reducing user trust.
• Existing fall detection methods require improvement to balance sensitivity with alarm accuracy.

Purpose of the Study:

• To present a novel approach augmenting traditional fall detection systems with a robot and Ultra-Wideband (UWB) tracking.
• To reduce false alarms from wearable sensors by using a robot for on-site verification.
• To enhance the reliability and user compliance of fall detection technology for older adults.

Main Methods:

• Developed a novel fall detection algorithm integrated with an alarm notification system.
• Utilized a UWB-based tracking system for real-time location of individuals and a home robot.
• Implemented a machine learning algorithm to compensate for multipath effects in UWB positioning for no-line-of-sight conditions.

Main Results:

• The fall detection algorithm achieved 99.0% sensitivity and 97.8% specificity.
• The alarm notification system demonstrated a maximum delay of 106 ms.
• The UWB tracking system, with the ML algorithm, achieved a positioning error of less than 0.2 m in various conditions.

Conclusions:

• The integrated system effectively verifies fall events, significantly reducing unnecessary emergency calls.
• The UWB tracking system is accurate and reliable for indoor localization, even in challenging environments.
• This novel approach enhances the safety and confidence of older adults using fall detection technology.