For one of my university projects I need to detect the presence of a person in the shower. In order to do that what I have in mind is use a proximity sensor. If the target is in a distance range (let's say 20 to 50 cm) It will assume that the person is in the shower when the distance measure is no longer in this range then it will mean that the person left the shower. I need to detect all the time of the person in the shower (meaning when he/she gets in and then out). I have read a lot of info and forums to properly choose the type of sensor. However I still don't have a clear answer. About the Ultrasonic proximity sensors I know that they are cheap and more accurate by it can be affected by Temperature, humidity and some ambient noise, all of them which happen in a shower. The other option an Infrared one gets affected by lightning conditions and the surface of the target. The last option I checked is a PIR sensor I know it senses Temperature of the target, but I don't know if it will be affected by the running hot water and also is more expensive. In general, I don't know how the running hot water will affect the detection of the person in shower. Also,so again I don't know how this will work. Sorry for the huge amount of questions, I am a little bit confuse with all the factors to have in mind. I am hoping someone out there can help a lost newbie like me and give me a suggestion. Thanks a lot :)
There is no one sensor that is perfect, but some are more reliable than others.
Yes, the temperature and humidity can play an important role in your sensing, and has to be taken into account. That's not to say that things like Ultrasound don't have a place in the scheme of things.
There's one word you need to think about here: Correlation.
If one single sensor isn't completely reliable then add a second that works in a different way. If both of them agree that there is someone in the shower (i.e., the sensors correlate) then the probability that there is actually someone in the shower is considerably higher.
The great thing about a shower is you are working in a defined enclosed space. That makes it much easier and there are plenty of sensors which, when combined with others, would be quite good at reliably sensing the presence of someone in that enclosed space.
Ultrasound above the person's head
- Simple to arrange.
- Can be fooled by the water from the shower head.
- Steam isn't good for them.
- Can be sealed
- Harder to work with
- Can be blocked by steam
Horizontal infra-red beam
- Can be positioned outside the shower glass with a reflector inside
- Small sensing area
- Can be blocked by steam
- People weigh more than water so is very reliable
- Simple to interface
- Hard to install safely
- Needs to be completely waterproof and sealed.
- Could be a shock hazard if not waterproof
You notice I haven't mentioned PIR. I'm not a fan of PIR for sensing the presence of humans. They work by detecting moving heat sources. If the person is moving (entering an area, walking round a room, etc) then they get detected. If the person stands still (as is often the case in a shower) it "forgets" they are there.
Of the choices above (and I am sure someone would come up with others) I would probably opt for the infra-red beam. Maybe split it into a V shape to cover more area - that is, the beam enters the side of the shower at an angle, strikes a reflector on the shower wall, and is reflected back at another angle. The detector is then placed away from the emitter. You could also have a second set to give better coverage.
Infra-red distance measuring in/on the ceiling could be used to improve correlation and thus the probability that there is actually someone there.
I'd stay away from pressure sensing unless you can get hold of a proper waterproof pressure sensing mat designed for in-shower use. Don't try and make one yourself, waterproofing isn't an easy task, especially when people stand on it all the time.