2

I am asking this rather open ended question. I am starting to explore IoT and its power to transform retail industry. Though I have heavy background in engineering, I am fairly new to IoT.

What I am looking for is an accurate indoor positioning system which I can organically develop - I do not want to use third party systems like Estimote or indoo.rs

I have two options at my disposal.

  1. Use Inertial Navigation System with a skeletal beacon based frames of reference (For sanity check and accurate positioning.)

  2. Use exclusively UWB Beacons for positioning.

I know most companies are going for option #2. Where I have problem with that is how inherently unreliable the Electromagnetic propagation is. I am not sold on the idea of measuring signal strength to determine distance. (A low battery power and a weak signal can show 10x greater distance). Plus maintainability of becaons is another issue.

With Option #1 there is always the issue of cost for INS and some say there is noise issues (especially with cheap INS systems).

So how does #1 compare with #2 for indoor positioning?

Improve this question
  • What's your target device to be determining the location from? e.g. what are you trying to track? – hardillb Mar 24 '19 at 21:03
  • @hardillb It will be a big four wheeled shopping car. Or to stretch it a shopping basket - the type which you haul using handles. – Ace Mar 24 '19 at 21:48
2

Inertial navigation will need active components on every object being tracked and the quality will be proportional to the cost of the components. It will also need to filter out every time it gets bumped into something. Errors with inertial navigation systems are cumulative so it will also need a way to reset location at known points. And as the basket is filled and gets heavier it's characteristics will change which will need to be taken into account.

Where as placing a beacon on the basket allows all the tracking equipment to be attached to the fixed assets in the environment and use triangulation from multiple receivers to determine location. Triangulation helps to smooth out things like battery level and RF occlusion especially as the number of receivers increases.

A third option is to use a computer vision based solution, retail environments tend to already have good CCTV coverage which can be used as a feed. Modern systems support multi camera input and can handle occlusion. This had the benefit of not needing any new infrastructure.

Improve this answer

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.