This requirement is looking for similar functionality to indoor tracking, but for use outdoors where there are no walls. i.e. Instead of people/assets inside walls for indoor-positioning, the assets would be outdoor but near a vehicle. e.g. I would envisage N * devices mounted on top of vehicles that would ‘together’ determine the relative position of assets within 20 or 30 meters of the vehicle.
Is this feasible using estimote products?
Also, I understand that distance calculations can be 20% out, but is that still the case when N * devices are used for position detection?
I think the major problem to overcome here is, the beacons on the vehicle will be rather close together, and if the distance-estimation-error due to Bluetooth signal is larger then the spread of the beacons, then suddenly using that to compute the position becomes very tricky.
Here’s an attempt to illustrate that:
20 m <===>
x <========> 1---2
Assume 1/2/3/4 are beacons, and that the estimated distance via Bluetooth can be up to 30% away from the real distance. Under these assumptions, worst case scenario is: for beacon 1, you would see an estimation of 26 m (20 m * 1.3), while for beacon 2 you would see 14.7 m (21 m * 0.7). Which may lead you to believe that the asset is actually on the right side of the vehicle!
That said, we’ve got beacons that also support UWB, a different radio technology that’s way more precise. Positioning using UWB is something we’re still experimenting with, so it’s best if you get in touch with our team at email@example.com to talk about that.
Thank you Piotr, that makes a lot of sense, and explains why indoors-on-walls works (anchors surrounding tags) while the opposite would be too inaccurate to be useful. Thanks again.