Hey guys! I’m currently looking into starting a ST robotics project. My current goal is to create an app that would have a virtual joystick, which would control a small remote control robot robot. That being said I have a few questions.
Has anyone ever used a hub to send arrays via JSON? Has it worked well? Is it efficient?
How much of a delay should one expect when dealing with a bot like this?
What methods have you guys used to send messages to the hub? This is currently my only major problem, as I have no idea how I will have the hub get messages from the app.
I’ve never used ST products for robotics but I have used Xbees, is it safe to assume that (as far as the code is concerned) they’re interchangeable? I have a few libraries I used to control a robotic via Xbee, and I want to be sure it should still work for the ST hub.
If anyone has any advice or would like to talk about the project in greater detail feel free to post below or email me at diazx069@umn.edu
Cheers!
Mike D
2 Likes
tgauchat
(ActionTiles.com co-founder Terry @ActionTiles; GitHub: @cosmicpuppy)
2
Fully respectful question, as it sounds like an interesting project, but…
Why SmartThings?
A lot of what you propose will not be easy to do on this Platform. And much easier with alternative frameworks and a few custom bits.
SmartThings programming relies on a stripped version of groovy. You are not allowed to add your own libraries.
You could run everything on something else and just use ST as a zigbee or zwave bridge, but you’ll be pretty limited in which commands you can send. Plus, mesh. So no forced sequencing and you’re pretty much limited to one cycle per minute.
I’m with @tgauchat on this, ST doesn’t seem like the best platform for robotics. But it all depends on exactly what you’re trying to do.
Mesh networks. If I tried to use anything else in my house it wouldn’t have enough range for the bot to explore the house, and with iffy wi-fi in the house using Particle products (or similar wi-fi based doodads) wouldn’t work. Besides i’m a ST ambassador, and i’m kind of interested in getting involved with harder projects.
OK, then will this be a network dedicated only to the robot(s)?
Zigbee can be a really inexpensive way to do robotic control, but usually you limit the network so you don’t have delays and sequencing issues. Typically the mobile robot and some sensors.
I don’t really see it mixing well into an HA network. But if it’s just for fun (or youtube fame), why not?
Yeah that’s what i’m kind of worrying about right now. There’s already quite a bit of lag as is with the ST things have right now. Even with extremely efficient code, I can’t help but feel as if the lag will be unbearable. We’ll see how it works now though. I’m debating simply using higher grade Xbees and seeing if they’re strong enough to work well indoors.