When can we get the UWB Location packet specs for non mobile based applications i.e. Raspberry PI

I’m interested figuring out how to get the location of a mobile platform with a raspberry pi and BLE onboard using your UWB location beacons. Is there a spec like (https://github.com/Estimote/estimote-specs/blob/master/estimote-telemetry.js) for the location information?

What I want to do is what the Indoor Location App does but without a mobile phone. Thoughts?

1 Like

There’s kinda two things we’re talking about here:

  1. Estimote Indoor Location is mostly an iOS/Android algorithm that takes packets from multiple beacons, and a known floor plan, and does a lot of math to compute an (x,y) based on that. It doesn’t use UWB at the moment, just regular BLE, we only use UWB to automatically generate a floor plan. We don’t currently have plans to bring it to non-iOS/Android platforms—but as always, let us know how you’d want to use it, our roadmap is heavily inspired by requests and use cases coming in from our community.

  2. The “UWB packet”, i.e., a packet which a beacon broadcasts that says what other beacons it can detect, and the distances to them (inch-level accurate, thanks to UWB), is something we’ll be publishing later this year. (Currenta ETA is later this Summer, but we make no promises ^_^) The caveat here is that UWB is not a low-energy technology, so it’s not running all the time and needs to be triggered, so for the packet to be useful to anybody we first need to build a public API to trigger it.

Hi,
Check out the Estimote Robotics-Indoor-SDK . It works with both RaspberryPi (Raspbian + ROS) and Ubuntu (Ubuntu + ROS) based solutions.

How can I get the UWB packets from UWB Beacons? Which API i Can use?

I have a similar need as you. Developing an inventory system for a manufacturing company. What I am trying (working on now) is running the AndroidThing OS on Raspberry Pis. Is nearly identical to the Android OS. The Pis will communicate with the beacons and send data to a JSON server. Maybe something similar would work for you.