I am wanting an alarm clock that will continue going off (buzzer, bed shaker, maybe light too) while the person is in bed and monitor the bed after the person leaves the bed for some time after so they don’t go back to sleep. The alarm clock alarm functions hardware and software will be easy enough. Sensing the twin bed for a person with an Arduino nano is where guidance would be helpful.
I am thinking about making pressure switches to put under all 6 of the legs on the bed frame, each set to about 50-75 pounds. Thinking about sandwiching rubber between 2 aluminum plates and having some sort of contacts in the middle. This will make contact when the rubber is compressed enough.
Inside the seat, you will find a pressure sensor, a silicone-filled "bladder," and an electronic control unit (ECU). When someone sits on the seat, the pressure sensor signals the occupant's weight to the ECU . The ECU then sends that data to the airbag, which has its own control unit.
A PIR sensor should see the warmth of a body, no? Far easier to integrate, less parts, and doesn't depend on the weight, just presence/absence of a radiant body.
Although, many of them are intended to sense movement, so if the person doesn't stir, you may get false negatives.
That's billed as a "static human presence detector" - as much googling as I can give it fails to confirm that it isn't just a Doppler shift radar unit, which means that keeping still will blind it to human presence.
So they are not claiming it works for static humans. Same issues and opportunities that come up with the PIR idea.
I use a PIR in tandem with radar unit to detect human presence in my lab, they each work well and report a bit differently; both def need movement to function.
Definitely worth considering if less invasive methods continue to fail. Although 90° rotation seems like it would be better to dump the person out without having the bed rotating over them and blocking them from getting up off the floor.
The seat sensors and similar bed sensors were my first thought, but the surface area and durability of the sensor are big concerns. The seat sensors are supposed to be installed on the metal seat frame and only monitor a 6” square area. The $40 bed sensor is more likely to sense the person with its surface area, but is expected to be replaced every 6 months due to wear with it between the mattress and bed sheets.
Overhead sensors would be simplest to install and maintain, but they expect motion and heavy blankets will interfere.
The force sensing resistors are an excellent suggestion. I found some with adequate max sensing weight for about $12 per pair on Amazon. Only putting one under a single leg is an excellent suggestion. Will definitely be looking into this further.
Thank you for sharing this. Knowing the design criteria for the puck will be very helpful. Will definitely need to plan out to protect the sensor from shear force damage.
