Fall Detection Using Active Reflected Wave Detector
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing fall detection systems using active reflected wave detectors face accuracy issues due to multi-path reflections, leading to phantom objects being misinterpreted as actual people, which can result in false fall alerts when multiple individuals are present in the environment.
Innovation Solution
The system determines the fall status of the nearest person to the active reflected wave detector as deterministic, disregarding non-nearest objects to avoid false alarms, and uses a classifier trained with multi-occupancy data to differentiate between actual and phantom targets, ensuring accurate fall detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If active reflected wave detector is used to detect multiple objects, then detection coverage is improved, but measurement precision deteriorates due to multi-path reflections creating phantom objects
Solution Approach 1:
The patent segments the detected objects by distance, identifying a nearest object and one or more non-nearest objects. The fall detection analysis is then applied selectively only to the nearest object, effectively segmenting the problem to avoid the phantom object issue while maintaining comprehensive detection coverage of the entire environment.
Solution Approach 2:
The patent applies partial action by focusing fall detection analysis exclusively on the nearest object rather than analyzing all detected objects. This selective approach reduces false alarms from phantom objects while still providing effective fall detection for the most relevant target, accepting that non-nearest objects are not analyzed for fall status.
2Quantity of substance
If fall detection is performed on all detected objects, then detection completeness is improved, but reliability deteriorates due to false alarms from phantom objects
Solution Approach 1:
The patent segments the set of detected objects into nearest and non-nearest categories, applying fall detection logic only to the nearest object. This segmentation maintains reliability by avoiding phantom object false alarms while still monitoring the entire environment for object presence.
Solution Approach 2:
The patent converts the harmful effect of multi-path reflections (which create phantom objects) into a beneficial distinction: by identifying that phantom objects typically appear as non-nearest objects, the system uses the presence of multiple detected objects to automatically filter out false targets, turning the reflection problem into a filtering mechanism.
3Measurement precision
If pendant with accelerometer is used for fall detection, then measurement precision is improved, but ease of operation deteriorates as person must wear the device
Solution Approach 1:
The patent replaces the mechanical accelerometer-based detection system with a wave-based detection system (radar, lidar, or sonar). This substitution eliminates the need for the user to wear or carry any device, significantly improving ease of operation and user acceptance while maintaining fall detection capability through wave reflection analysis.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enhances the accuracy of fall detection by focusing on the nearest person and ignoring non-nearest objects, reducing false alarms and improving the reliability of fall alerts in multi-person environments.
Implementation Method 1
controlling an active reflected wave detector to measure wave reflections from multiple people to receive measured wave reflection data
Data Source
AI summary
One embodiment relates to a computer implemented method for detecting if a person has fallen in an environment comprising multiple people. The method comprises controlling an active reflected wave detector to measure wave reflections from the multiple people to receive measured wave reflection data that is obtained by the active reflected wave detector. The method further comprises detecting if a person has fallen in the environment using at least a portion of the measured wave reflection data, wherein a fall status of a nearest person in the environment, being a person of the multiple people that is nearest to the active reflected wave detector, is deterministic of a detected fall.