Presence / Occupancy vs Motion

I noticed with the ST motion sensor that if I stand still (and not really that still) it thinks motion has stopped…even with it right in front of me on the counter. This got me thinking…are there any zwave presence sensors? I don’t mean the fob. I mean a true occupancy sensor. One where I can sit on the sofa, watch a movie, and it’ll know I’m there even if I don’t move for a while.

I think everyone I have seen so far are all PIR sensors.

I do know the Aeon Multi Sensor has a sensitivity dial that can be set so high a flashlight will set it off.

A lot of newer HA stuff I see on Kickstarter does triangulation to determine presence. Basically you have three or more “base stations” in your house and then carry around your phone or some other device. Then based on signal strength to each base station they determine which room you are in. The downside of this is that you need multiple base stations of course, and need to carry around the device that is communicating with these stations.

I also wonder how accurate these would be. It would seem to be that walls, floors and furniture could really play havoc with signal strength. I suppose if calibrated first it could be fairly accurate, but then I wonder how much interference from other devices or just EM noise might effect things.

I’m actually looking for it to just recognize that people are in the room…guests for example. Like I said, I noticed that If I’m in my kitchen reading something and not moving, the sensor goes into no motion mode. Now, I can get around that by increasing the timeout. However, that is a bit of a kludge. I’m looking at it more from the watching a movie aspect. If you are sitting on a couch watching something, you tend not to move much. (Obviously this is a first world problem!)

@brianlees I wonder if you could use one of the wearable devices like an UP or FIT to tell the room that you are still moving… as even if you are reading your hand would move every now and again… I know this is a huge reach but a thought towards something in this direction, the motion sensor realizes you are in the room and that a now wearable device is in motion so it keeps it on??? just spitting concepts/ideas out there

@tail24 I doubt it. Kind of defeats the purpose of little to no interaction. :smile: However, this thread was just a question. I’m not hung up on getting them now. I noticed the issue and I was curious if there was any solution. I think my solution is to find the most sensitive detector for certain areas and increase the timeout. @urman mentioned the Aeons were really sensitive. So, that might be a solution for a few areas. Hopefully, ST will get that device database up and running so people can leave comments about devices like this to help others! :wink:

Well, it’s not in the database, but… I have two of these devices and I think they work well. I have one in my upstairs hallway that it’s tied to a nightlight app for when someone gets up to use the bathroom in the middle of the night. I also have one in my den that I hook up to me Xmas tree in December. It does a pretty good job of detecting motion but there were times if you sat for too long that it would ‘time out’ on us. Usually just waving a hand in the air (like I just don’t care) quickly turned on the tree again.

yup Aeon’s works great, but it does fail when you’re stationary.

I have one setup in my living room facing my couch. If I’m reading or watching tv with little to no movement it’ll think someone left the room. The way PIR sensors works is that it senses changes in the infrared signal. So once it detects motion, it’ll keep a time counter until the infrared signal changes, and resets the counter. When you stand still, it senses no change in the signal; the timer will eventually lapse (based on the configuration of the sensor) and send a no motion event to the hub.

Increasing the sensitivity will increase the affective range of the sensor, but also pick up smaller movements.

When it fails, just give your TV a good wave to let it know you’re still there. Think of it like how Pandora bugs you when you don’t change tracks for awhile.

So I’ve tried the ca9000 and enerwave ceiling mounted sensors. The ca9000 claims to be an occupancy sensor but is basically the enerwave with a built in timeout.

Any progress on something that actually detects occupancy vs motion?

Curious if there has been any more product development here, as I see this as the repeat of a big shortfall the lighting industry had with motion sensors 15+ years ago. The ones like that which replaced the wall switch in the bathroom that people just ended up tampering with cause they kept turning off the lights when you were on the John. That industry had to re-brand and re-sell motion sensing as “occupancy sensing” to make it take off again, and the technology is now found in many commercial buildings.

I just hooked up my first ST Motion Sensor and am having these same issues with motion lacking in my office. I want the lights to come on and stay on while I am in the room (even when sitting still at the desk), but then go off within seconds of the room becoming vacant.

TL;DR - Are there now any occupancy sensors on the market that integrate with SmartThings?

Reviving this topic, because I see it as a gap too.

what if you share a garage with three others, and the first one in is the one that gets the spot (duh lol). But you don’t want the others to have to open the garage when pulling into the driveway to determine whether another vehicle is there, thus putting wear and tear on the opener and system. You don’t need to know WHICH car is in the garage, all you need to know is whether the space is occupied by a vehicle or not.

A pressure-sensitive mat might be the ticket, unless the vehicles have different lengths/widths. If one is a Miata and the other a pickup truck, they will likely not hit the mat at the same place.

A typical photocell arrangement might work, except then you have to ensure no one accidentally blocks the beam with a piece of Sheetrock or a shovel or something.

So I’m imagining a ceiling mount motion sensor that also includes some sort of ‘sonar’ or other rangefinder. A plug-in, not battery operated device. When it detects motion, it takes a range reading. Below a certain threshold, it says “occupied” - and above that threshold, it says ‘vacant’. And whatever that reading is remains in-place until the next time it takes a reading… which is the next time it detects motion.

Seems this would be pretty straightforward (and inexpensive) with the range sensors made for robots.
Might have been an interesting candidate for the now-defunct ST Shield.

You can do this now by hanging a tennis ball on a rope, hanging it from the garage ceiling, and putting a tilt sensor either on the rope or inside the tennis ball. :sunglasses: :construction::blue_car:

A lot of people use the hanging tennis ball anyway as a marker to tell the driver where to stop the car as they’re coming into the garage. This just gives you a networked indicator that the car is there.

And there was a community member who built a car detector, let me see if I can find that one.

Here you go: this one is a lot more complicated than just the tilt sensor tennis ball method, though. :sunglasses:

I actually tried the tennis ball approach, in low-tech, a few times over the years. Somehow, it never worked out lol

1 Like