Detecting Human Falls with a 3-Axis Digital Accelerometer
Elderly individuals can suffer accidental falls due to weakness or dizziness. The initial injury can be further aggravated if treatment is not obtained within a short time. MEMS-based fall detectors can sense changes in body position by tracking acceleration, determine when an individual is falling, and issue an alert for assistance. This article describes the use of an ADXL345 three-axis digital accelerometer as a fall detector.
posted by Analog Dialogue @ 3:39 PM
3 comments
3 Comments:
Human falls are very common not only to the elderly people but also to anyone who has muscle imbalance or who sufferred a pre-existing desease which results to postural incoordination. A very healthy person is also experiencing fall in their everyday activities, this Accelerometer will help us to prevent broken bones, nasty cuts,head traumas, splints/braces,heavy cast and even death. It will send us signals or cue for our body to make a preparation and corrective action to prevent sudden fall.
Im sure as soon as everyone realized the benefits of prevention before further damage is produced they will grab a 3-axis
Digital Accelerometer for themselves and their family
This article could have been titled 'I've fallen and I can't get up' - a famous commercial for a pendant with a button to call emergency services. But one important difference is that the injured person doesn't even have to be conscious to press the button with this idea.
And the required external processing power is truly negligible. The author cleverly strings together some of the built-in interrupts available in the xl345 to construct a fall algorithm. Reading the series of interrupts could be done by a processor as simple as one that can be found in a digital watch. In fact, the accelerometer could be in a watch, eliminating the need for a pendant or some other special-purpose fall detection device. This makes it a more attractive solution to an aging relative.
Where are the references to the original research about falling. This would be very usefull.
I am slightly sceptical that this all that is required to detect a fall. Are all falls similar? How do you test this?
Thanks
Post a Comment
<< Home