Motion Sensors for Occupancy Tracking

This NewScientist article from last spring shows a very cool system of motion sensors for occupancy tracking. The sensors are installed more densely than in typical buildings, so they can detect occupants’ positions fairly precisely within hallways.

The most interesting part of the system for me is the video showing playback of motion through a building — you can clearly see where people were walking, where they ended up, etc.

This seems like a natural fit for a smart house — instead of just watching where there’s motion now, pay attention to where the motion has gone and remember that there must still be someone in that room (even if they’re not moving much now).

$10 Parallax motion sensors + blinders to narrow the cone of coverage + write open-source software for the capture and playback, anyone?

4 Responses to “Motion Sensors for Occupancy Tracking”

  1. Phil Winder says:

    I’m very interested in this application. I had invented a new type of proximity detector exactly for this purpose for my masters thesis, but trying to squeeze some commercial money out of it is hard.
    So I have been considering just publishing it anyway, maybe writing a paper, but then its out there for anyone to steal.
    Have you had any experience like this?
    I really don’t mind information flowing into the community, I just dont want anyone to make some serious money from this just because they have more contacts. Is it worth holding on to it for longer?

    Cheers,
    Phil
    P.s. Avid reader, great site!

  2. Dave says:

    The only real options are to either patent the idea, publish the idea, or sit on it.

    Patents are expensive to produce (since you almost certainly need either a Patent Attorney or Patent Agent[1] to do the paperwork [1], and these aren’t cheap), and the only thing they really give you is the right to sue if someone infringes on your idea (Of course, hopefully, any company interested in the idea would license it from you rather than infringing it. On the other hand, the vast majority of patents are never licensed, nor even used.).

    [1] It is theoretically possible for an inventor to write his own patent. However, I strongly urge against this, since the language must be VERY precise, else you end up patenting something entirely different from what you intended (e.g., You may fall flat on your face when you try to file an infringement suit.).

    [2] I’m neither a Patent Attorney nor a Patent agent (nor even a regular non-Patent Attorney), but I do have a handful of patents, so I have been through the process a few times.

    You can, of course, go the publication route. But, this puts the idea into the public domain, and anyone can then use it (without paying you royalties). But, you do have the prestige of having public acknowledgment of the idea.

    You can, also, sit on the idea. However, if someone else comes up with the same idea, well, you don’t get any benefits (and can even be locked out of using the idea if they happen to patent it [3]).

    [3] The only prevention you’d have is if you could prove prior art, but proving prior art would be impossible if you hadn’t revealed the idea.

    Anyway, interesting idea about the room tracking system. One flaw, though, may be if more than one person is in the building. If two people enter a room together, and then one leaves, will the controller get confused and turn the lights out on the person remaining in the room?

    It’s little glitches like this that make most systems impractical. I’m aware of some industrial/commercial outfits that use motion sensors to control when the lights in a warehouse type of building should be on (and which “zone” that they should be on it, but these use simple motion sensing, and depend upon the person moving within a 15 minute period or so, else the lights get turned out.

    Dave

  3. Keith Neufeld says:

    Dave, the complications for human tracking are all a matter of software. :-)

    I see motion tracking supplementing motion sensing in a smarthome implementation, not replacing it. That is, if someone goes into a room and becomes stationary, the system should remember that they must still be there, whereas a motion sensing application wouldn’t know they were still there.

    If someone else wanders through and after they’re gone there’s no movement in the room, the system doesn’t know whether someone is still there or not — but neither would a motion-sensing-only (without motion tracking) system. Adding motion tracking (correctly) shouldn’t make things worse, but nor is it a cure-all.

  4. Phil Winder says:

    Thanks for the thoughts Dave!

Leave a Reply