Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Privacy-Preserving Bathroom Monitoring for Elderly Emergencies Using PIR and LiDAR Sensors

Youssouf Sidibé, Julia Gersey

TL;DR

In-home elderly care requires reliable detection of falls and prolonged inactivity without compromising privacy, especially in bathrooms. The authors implement a privacy-preserving monitoring system that fuses PIR and LiDAR data with a rule-based state machine to detect Entered, Seated, Exited, Fall Suspected, and Alert states, with data sampled every 50 ms. The approach uses a low-cost hardware stack (under $20) and validates its effectiveness through five real-world experiments, demonstrating accurate state transitions and timely alerts while avoiding video capture. This non-visual, passive design supports aging-in-place, reduces alarm fatigue, and offers a scalable, explainable solution suitable for deployment in multiple rooms and homes.

Abstract

In-home elderly monitoring requires systems that can detect emergency events - such as falls or prolonged inactivity - while preserving privacy and requiring no user input. These systems must be embedded into the surrounding environment, capable of capturing activity, and responding promptly. This paper presents a low-cost, privacy-preserving solution using Passive Infrared (PIR) and Light Detection and Ranging (LiDAR) sensors to track entries, sitting, exits, and emergency scenarios within a home bathroom setting. We developed and evaluated a rule-based detection system through five real-world experiments simulating elderly behavior. Annotated time-series graphs demonstrate the system's ability to detect dangerous states, such as motionless collapses, while maintaining privacy through non-visual sensing.

Privacy-Preserving Bathroom Monitoring for Elderly Emergencies Using PIR and LiDAR Sensors

TL;DR

In-home elderly care requires reliable detection of falls and prolonged inactivity without compromising privacy, especially in bathrooms. The authors implement a privacy-preserving monitoring system that fuses PIR and LiDAR data with a rule-based state machine to detect Entered, Seated, Exited, Fall Suspected, and Alert states, with data sampled every 50 ms. The approach uses a low-cost hardware stack (under $20) and validates its effectiveness through five real-world experiments, demonstrating accurate state transitions and timely alerts while avoiding video capture. This non-visual, passive design supports aging-in-place, reduces alarm fatigue, and offers a scalable, explainable solution suitable for deployment in multiple rooms and homes.

Abstract

In-home elderly monitoring requires systems that can detect emergency events - such as falls or prolonged inactivity - while preserving privacy and requiring no user input. These systems must be embedded into the surrounding environment, capable of capturing activity, and responding promptly. This paper presents a low-cost, privacy-preserving solution using Passive Infrared (PIR) and Light Detection and Ranging (LiDAR) sensors to track entries, sitting, exits, and emergency scenarios within a home bathroom setting. We developed and evaluated a rule-based detection system through five real-world experiments simulating elderly behavior. Annotated time-series graphs demonstrate the system's ability to detect dangerous states, such as motionless collapses, while maintaining privacy through non-visual sensing.

Paper Structure

This paper contains 19 sections, 6 figures.

Table of Contents

  1. Introduction
  2. Background and Related Work
  3. System Design
  4. Hardware Setup
  5. State Logic
  6. Experiments and Results
  7. Experiment 1 – Normal Entry + Toilet Sit
  8. Experiment 2 – Long Sitting (No Motion 10+ min)
  9. Experiment 3 – Entry + Immediate Exit
  10. Experiment 4 – Simulated Fall (Out of FOV)
  11. Experiment 5 – Collapse While Entering
  12. Discussion
  13. False Positives and Negatives
  14. Why This System Matters
  15. Future Improvements
  16. ...and 4 more sections

Figures (6)

  • Figure 1: Experimental sensor setup in a test bathroom.
  • Figure 2: Experiment 1: Entry, sit, exit. Correct transitions detected.
  • Figure 3: Experiment 2: Alert triggered after prolonged seated inactivity.
  • Figure 4: Experiment 3: No alert triggered. System correctly classifies short visit.
  • Figure 5: Experiment 4: Fall suspected after distance spikes and no motion.
  • ...and 1 more figures