Use Case Feasibility - Swarm Relative Ranging

[iQuad427]

Hi,

I'm currently working on creating a relative localization system in a swarm of robots, and I'd like to discuss the possibility of using this repository as basis for that.

The system would work by putting both an initiator and a responder node on each robot. Then through the use of this repository, I would be able to recover range to/from all other robots by targeting all responder with each initiator.

Since I never used Zephyr before, it's a bit difficult for me to get started with the repo. That is why I'd like to have your ideas on the feasibility of the project before going head down into it. Also, if you have any advice on how to get started, I'm all ears.

Many thanks in advance for taking the time to help me!

[wassfila]

Hi,
I'm happy to see that this repo is inspiring you and that you consider using it in projects.
Ranging of all points across each others of moving targets is indeed the most challenging problem.
I'm personally no longer active in this project as it was for me a learning experiment, and I defined a design I wish I see in other professional libraries that will be officially supported.
My conclusions were that UWB is so complex that it kind of require considerable investment and I did not see many opening that up to open source. If you have a high stakes project with high budget, it is better to hire a company. For research projects, you can try with what is available but without guarantees.
You can use and reuse, modify any part of code I have created without any issue, but in this repo I had to inherit the license, you might have to consider my dependencies licenses mainly zephyr.
I would highly recommend to first get fluent with Zephyr which is inspired from Linux kernel, which is quite hard core to be honest, I encourage you to do it, it's not hard, just has to be taken serious as its own task and not as side thing.
In embedded microcontrollers programming Zephyr is the most professional env I know for large scale projects. The only case I would not recommend it is if you have a supplier giving you a ready env in e.g. Arduino or other where you only modify the application. But if you need to play with timers and close to hw functions, you might need 100% control and there you're already beyond the os but rather focus on your hw peripherals and registers capabilities.
This project was tricky as I tried to connect high level python through json with low level precise timing functions, and that approach is something I recommend using this or any other library, so that you work with higher level abstractions.
Good luck.

[wassfila]

this is the higher level python framework I was referring to https://github.com/HomeSmartMesh/mesh_position
and this is my doc https://www.homesmartmesh.com/docs/networks/ultrawideband/ it's been a long time since I made the survey in there, I hope it's better now.