So, I’ve done many hours of reading, and have played with a few of the dev boards with mixed success. I’m working with a small team that’s looking to implement a basic indoor asset tracking system. I’m mainly interested at this point in finding someone who has implemented a system that tracks a hundred or so devices with like 5-10 “anchor” nodes. I’d be up for paying a small consulting fee for like 30-60min on a zoom call to answer questions:
What existing RTLS firmware is out there now? PANS is retired/for DW1000 only?
What are most using for host devices for “backhaul” of the RTLS data?
General Q&A.
Our goal is to get a proof of concept of some kind within the next month or two to see if our clients like what we could deliver. I have several DWM3001CDK.
If PANS doesn’t do what you need a basic RTLS isn’t necessarily that hard to put together depending on the requirements. I mean it’s hard as a hobby / student project but shouldn’t be for professional development team. The difficulty depends a lot on what you actually need to do.
Key questions that dictate a lot of the difficulty:
What is the accuracy requirement? +/-10 cm 2d position most of the time is fairly easy, significantly better than that gets harder, good all the time rather than most gets a lot harder. 3d position accuracy is a lot harder, the geometry for most real world installations doesn’t lend itself to good height accuracy.
What is the power requirement? UWB may not need much power to transmit but you need to work out when to transmit. Figuring this out can easily take more power than the actual transmission. The more devices and the higher the update rate you are aiming for the tighter this timing needs to be.
How large is the area and what is the environment like? A big empty box is easy (but not too big or you hit range issues). Lots of metal and obstructions is hard, especially if you have a tight accuracy requirement.
What are the latency / update rate requirements? What needs to know the locations? This has significant system architecture implications. If the tag needs to know it’s location with a low latency then that is a very different system to a server in the building needing to know where a tag was at some point in the last minute or two.
Any restrictions on the infrastructure? Can you run special timing cables to the anchors on the wall? Ethernet? Power?