Ask HN: Indoor Location Tracking?
I want to provide point-to-point indoor/outdoor wayfinding in a way that works at scale (for our largest customers, hundreds of global locations).
One of the challenges is figuring out where the person is within the building.
It seems there’s two sub-optimal options:
1. Use Bluetooth beacons. Involves purchasing & placing many beacons. High accuracy but also high effort. Works with Android & iOS
2. Use WiFi (RSSI). Leverages existing infrastructure. Relatively accurate, with relatively low effort. Works on Android, but involves additional work and perms with iOS (including sending our customers’s mapping data to Apple?)
Other methods using light and sound exist, but are also “high effort”.
Anybody have some insight I’m missing, or an idea for a novel approach?
I’m open to collaborating on this!
Have a look at - https://www.hyperar.com/
I am not affiliated in any way. Just found it a couple years ago.
Thanks! I filled out their “Early Access Program” form a couple weeks ago but haven’t heard anything back.
Using the phone’s camera to pinpoint a person’s location does seem like it would work well, but getting the image data of all these building’s floors (as well as responding to changes within the floors) seems like a huge challenge
I used to work for a large museum a few years ago (2020-2021ish ish) and we looked into this quite a bit and came to the same conclusion as you... none of the methods were very easy or foolproof. Sounds like you've already done some research, but the keywords that were helpful for us were "indoor blue dot positioning" and "indoor SLAM (simultaneous location & mapping)".
The UX was also an issue. I don't think many normal users really like having to install yet another buggy app just to wayfind around a venue they might visit only once a year or once in their life. And making it work without a proprietary app (i.e. in Google Maps or Apple Maps) was much harder. In our visitor surveys, there was pretty much universal disdain for this sort of imposed technological solution; nobody really wanted it. And that's assuming they could even successfully use the app; a lot of the solutions (light, ultrasound, Bluetooth, etc.) require users to manipulate their device in a way that they don't typically associate with location. Unlike GPS where they just wait a few seconds for the circle to shrink, the proprietary apps depended on some sensor not really meant for wayfinding (the camera or mic or bluetooth stack), and all of those things vary a lot between devices (especially Androids). It seemed like it'd be absolute nightmare to maintain and debug across real-world devices.
In our case it was a self-imposed problem... we used to have paper maps, but management became scared of Covid particles and banned those and demanded that we come up with a digital solution instead. We pushed back hard, but lost, and were forced to brainstorm a digital solution instead.
So we did our research and presented our findings, but ultimately we just gave up on the whole project. It wasn't worth the expense and complexity, given that the field was still so immature and probably whatever we adopted would be obsolete and unsupported within a few years. There were a lot of vendors hawking their expensive proprietary solutions of dubious effectiveness, and we couldn't be assured that any of them would still be in business in a few years.
In its place we put up a bunch of signs with different QR codes. Much simpler and cheaper.
Separately, though, I'm reminded of my experience with UWB (ultrawideband) Airtags with a recent-model iPhone... THAT worked phenomenally well. It was kinda magical to experience. But I don't think you could easily scale that up to a mesh of Airtags scattered around a building, at least not out of the box. And many Androids don't have UWB radios at all or don't have the software support to enable them yet. Google is way way behind on this =/ They finally launched a few this year with some partners, to middling reviews.