Sam Asano's Let's Invent: Finding a solution to detecting fallsBy SAM ASANO
May 13. 2018 7:21PM
We are entering a phase of invention in our falling down detector project called “you better come up with your solution.” A large number of amateur inventors start dropping out around now.
The reason seems to be that they are impulsive rather than step-by-step thinkers. Developing a solution to a complex problem can be time-consuming and sluggish and outright boring.
- - - - - - - - - - -
A fall can take all sorts of forms.
Falls can occur from a momentary blackout for various medical conditions. The fall is extremely injurious as the person is falling down unconscious and has no means to deploy any counter movement while falling. Most people who fall this way do not remember what happened to them.
A fall takes less than one second, and unless you are young and physically fit, there isn’t much one can do to reduce the impact of the fall by extending your hands to support your falling body.
I have amply explained and emphasized to you that a failure-free automatic reporting system is an essential development we need.
Now, the next stage is to decide where the sensor sensing the fall should be installed.
This sensor or sensors must be able to sense a fall of the wearer and by some means also be able to transmit to the outside world the fact the wearer has fallen down.
One possible solution is to install a number of pressure sensor in the clothing the person wears. Since the pressure sensor would register zero pressure when the person is standing up normally, the fall signal would be transmitted when one or more of the pressure sensors would register positive pressure by the person lying down on the ground.
Since there would be plenty of space in the cloth to install and hide the transmitter and rechargeable batteries, manufacturing this system would be easier. The main issue is whether the system embedded in the cloth would tolerate the wear and tear of people wearing the cloth and occasional washing. Another serious issue is that the wearer needs to have both upper clothing and pants in order to make sure the fall is detected without fail.
One more problem on this solution is that the head of the wearer is not covered. We cannot demand that the person wears a helmet, we just have to leave the head uncovered.
Develop a tiny mechanical gauge that detects the vertical angle of the person, and if the person has fallen down, it will transmit the alarm by detecting the person is no longer standing up. This solution is basically more attractive to consumers as one needn’t have more than one such gauge attached to his/her body. Perhaps such a gauge can be sewn onto underwear with easy access to change the rechargeable battery. Again, the gauge/transmitter combination must be reliable and able to take wear and tear of the daily use. This concept is being developed by our firm at this moment. However, miniaturing is a difficult issue.
Add on a feature to a wearable watch in a form of an app, that detects a fall by analyzing accelerometers and gyroscopes built in the watch body. This principle is easy to conceive, but I see a lot of difficulty. This is a noncontact measurement method and has a large potential of transmitting false positives and false negatives. As I stated before “false positive” is not as injurious to the wearers and the help desk as it is reporting an event that didn’t occur. But a “false negative” is very injurious to the credibility of the system. It is not reporting the event when it did occur. That is the weakness of a sensing system like this one.
Now, readers: Please develop your own sensing system(s) solutions. If you feel your system is sound, please send in the description (but after your careful thinking, please!)
The last week of a month is when I publish various reader comments and their invention ideas. Please send in to firstname.lastname@example.org or email@example.com. Please indicate the town you live, paper you read and phone we can reach you (optional).
Shintaro “Sam” Asano of New Castle was named by MIT as one of the 10 most influential inventors of the 20th century. Write to him at firstname.lastname@example.org.