Robot communications

[Kevincav]

I know this may not be the right place for it but I'm building a fully autonomous robot that will traverse through any building and map out a blueprint of what's in there. (Original idea was three quad copters, but seemed a little dangerous with all those props flying around people )

I'm looking at getting radio's for it, my original idea is an omnidirectional antenna on the robot that transmits outside the building to a repeater. Then attaching a unidirectional LOS antenna pointing back towards our computer. The kind of data being passed back would be a live video and coordinates. I was wondering if this seemed like a good option and if so what antenna's do you guys suggest for this?

Thanks for the help guys.

 

[Kevincav]

Also we're thinking about making the receiving end a raspberry pi to keep it small, so hopefully small antenna's are an option.

 

[KJ4CCH]

well if you are going to use video, then use a low band type video transmitter. (900mhz/1.3ghz, or even 443 mhz video) this may help with penetration threw objects. this would be stand alone compared to you data transmitter. then maybe use these for data:

http://www.adafruit.com/products/128?gclid=CMuov8_go7sCFWRk7AoduXoAkQ
here is a tutorial that i have not read threw to maybe give you an idea of how to use it with the Pi:
http://www.cooking-hacks.com/documentation/tutorials/raspberry-pi-xbee



for antennas, use a omni directional for the robot and input side of the repeater, then a directional antennas for both the TX of the repeater and a directional for rx at your station.

paint picture:




now, i am interested in the code you are using, and your sensor arrays! please tell more about your project!What algorithm are yall using? you are gonna need more computing power at the robots end, an arduino or maybe a raspberry wont cut it! (i dont think so, unless you do some amazing memory management!)

I hope i helped hahah

 

[Kevincav]

Thanks for the help KJ4CCH. As for as the code goes I'm using a depth-first search w/ backtracking. The robot itself has a full onboard computer so that'll help with the power, as far as the memory management goes, I'll be writing everything in C in order to manage the memory myself . The laser is using this laser range finder.

The client side is going to probably construct blueprint from given coordinates, and upload them to the server's Database (postgresql). The problem is I want to live stream everything back and let the user select a point, and get a path directly to that point. This would be using the A* pathfinding algorithm.

I haven't proofread it yet but if you want to take a look at this, it's everything I'm thinking of so far.

 

[KJ4CCH]

Thanks for the help KJ4CCH. As for as the code goes I'm using a depth-first search w/ backtracking. The robot itself has a full onboard computer so that'll help with the power, as far as the memory management goes, I'll be writing everything in C in order to manage the memory myself . The laser is using this laser range finder.

The client side is going to probably construct blueprint from given coordinates, and upload them to the server's Database (postgresql). The problem is I want to live stream everything back and let the user select a point, and get a path directly to that point. This would be using the A* pathfinding algorithm.

I haven't proofread it yet but if you want to take a look at this, it's everything I'm thinking of so far.

wow sounds pretty cool! good luck with your project! sounds like fun!

 

[rockets4kids]

If the building already has a wi-fi mesh, it would be silly not to just use that. A $10 WiFi dongle on you Pi and you're done.

 

[Kevincav]

Well the idea is that it can be used by emergency personnel, so we assume that the building won't have wifi.

 

[Kevincav]

wow sounds pretty cool! good luck with your project! sounds like fun!

Thanks man, I appreciate it.

 

[rockets4kids]

In that case, you are going to want to want to use the lowest frequencies possible for maximum penetration. Bandwidth is really limited at the lower ISM bands, so you probably want to investigate your options in the emergency bands.

 

[Kevincav]

In that case, you are going to want to want to use the lowest frequencies possible for maximum penetration. Bandwidth is really limited at the lower ISM bands, so you probably want to investigate your options in the emergency bands.

Yeah that's what I was thinking. I remember there being certain VHF freq's that were reserved for
emergency use only. I figured having something like that, obviously no on that range though .

 

[KJ4CCH]

well maybe look into the ham band? it is open for communication and experimentation. you only need to follow a few rules hahah. 70cm has video transmitters as well. We use the 70cm for our Long range RC links.


Hey just an idea being how you are into robots as well as i am, what do you think about open source robots under 100USD? i was thinking about starting a small company producing kits based around open source software to run them. I mean i google open source robotics ans such, and some of the cheap kits are in china, or over 100 bucks.

 

[Kevincav]

well maybe look into the ham band? it is open for communication and experimentation. you only need to follow a few rules hahah. 70cm has video transmitters as well. We use the 70cm for our Long range RC links.


Hey just an idea being how you are into robots as well as i am, what do you think about open source robots under 100USD? i was thinking about starting a small company producing kits based around open source software to run them. I mean i google open source robotics ans such, and some of the cheap kits are in china, or over 100 bucks.

Thanks man, I'll look into those. As far as the open source robotic shop goes, I think that'd be really neat. I would have loved to see a site like that when making this project. I'm not actually into robot's as much as you think I am, this is kind of my way of getting into them. I've always been trying to find a way to really start using robots and fpv droids, but haven't had the chance before this.