Slashdot Mirror
Bluetooth for Homebrew Robots?
MacGod asks: "As an extra-curricular for my university, I am working some other engineers to try and design a RoboSoccer Project (please don't wipe out our server if possible), with the hopes of entering the RoboCup tournament in the near future. We are currently using some low-frequency Abacom RF communication modules, but would like to able to go to BlueTooth if possible. However, I have found it nearly impossible to locate any suitable BlueTooth modules. We'd really like something that could be put onto a breadboard, and that would come with sufficiently detailed instructions that we'd be able to implement it without being BlueTooth experts. Any suggestion on homebrew BlueTooth solutions, Slashdot?"
Bluetooth modules for use with mobile robots
comp.robotics.misc has a few useful posts.
this post's author could be someone to ask?
Google's froogle feature is pretty cool, I found board-mountable bluetooth modules in less than 10 seconds (I had to resort by price).
Here's a Site that sells the modules.
Black and grey are both shades of white.
We are currently using some low-frequency Abacom RF communication modules, but would like to able to go to BlueTooth if possible.
Why Bluetooth? A better question would have been, "We are currently using some low-frequency Abacom RF communication modules, but would like increased range, bandwidth, and fidelity. Does anyone have any recommendations for a wireless technology which will meet these needs?"
Let the problem dictate the solution. It sounds like you're just itching to make use of Bluetooth because it's buzz-compliant.
Um you probably should have though of that before you post to /. That is like waving a red flag to a bull. Good luck!
It said "windows 98 or better" so I installed Linux
Here an interesting project that uses Bluetooth. Lego Robot.
Also here is a good overview of Bluetooth. Overview.
If you're having trouble finding something to use, maybe it's not the way to go......like other posts state, the range and BW are limited.....I'm using a nice little device from Honeywell - ROC09352XMS which is a radio-on-a-chip device for my thesis... All you need is a microcontroller with an SCI interface (i.e. just about any PIC or 68HC11 device, which most people have some knowledge of), a power supply and an antennae - everything else is self contained - and it can range from 300-900Mhz, up to 19.2kBps.....at $26US a piece, not very expensive either.....
Hello,
;-)
0 .h tml
p ro _wlsc.html
There do exsist small bluetooth modules, but not for what you want really. Bluetooth is fairly complex, and for small robots you'd be far better off designing your own protocol. And its more fun that way
But if you are dying for bluetooth, Google is your friend:
http://www.national.com/appinfo/wireless/LMX982
http://www.wirelessfutures.co.uk/products/prod_
I think there are others out there - I've heard of them before.
Have you though about using microwave communications? 28GHz ROCKS!
we, believe it or not, tried bluetooth for our vision->robot communication. my advice: try something simpler to just get your basic system running.
.. and drop us (site) an email if you have specific problems..
/me watches this comment be ignored.)
for example, we used Radiometrix RPC's for 4 straight years.. they are really limited but everyone uses them because their problems are well known in the league. all i can say is, good luck
-sergei (cornell robocup 2003,2004)
(sigh...
I'm doing some work with Bluetooth, using the CSR modules, "Casira" programmer, and the Bluecore programming tools. For simple stuff like headsets and RS232 links, you can run all the necessary code on the module.
For stuff based on CSR silicon, if I were doing things over, I'd go with the Siemens dev kit for their Siemo or Siemo2 modules. Same tools, more or less, but you're working with modules right away that you can spec into a final product. I didn't find them until I was looking for prequalified modules based on CSR's chipset...
Raise you're hand if you're one of those people who clicked the link to see whether the server had been swamped yet.... come on, be honest... that's what i thought.
It's not stupid. It's advanced.
Personally, I wouldn't. Wifi has outrageous power demands (as is seen by how quickly it drains my laptop batteries), and increases the requirements of the host processor. For big robots that can afford an onboard laptop, this might be fine. For smaller ones running on PC104, you might manage. For microcontroller-based robots, forget it.
Hardware, software, and blinking lights!
if you don't want your server being hit, don't put a link to it on /. front page !
Better yet, if your page doesn't have a lot of large images, link to a Google cached page instead.
Will I retire or break 10K?
I have some pictures of the board on the coure web site. The bluetooth module is just above the "Rev B" sticker in the first picture.
just having a few products doesn't mean much. Try doing a little research and you'll find the Bluetooth spec support 3 classes/ranges:
Class 1 = 100m
Class 2 = 10m
Class 3 = 2m
with corresponding power demands. All of which are still close to 1/4 the power of 802.11 devices.
LoB
"Anyone who stands out in the middle of a road looks like roadkill to me." --Linus
Depending on your platform, it may be a lot easier to use 802.11b. I've built mobile robots around Ipaqs and Geode single board computers. Since both have pcmcia slots, adding wireless is simple.
"The danger is not that a particular class is unfit to govern. Every class is unfit to govern." - Lord Acton
I think they can. In fact I hope that they kick ass and take names.
This is an academic contest. All of the teams are students and grad students, each sponsored by faculty members like them. There is a lot of work to do for this contest, but all previous work done by all previous teams is open to anyone who wants to join in.
My team gives away our design and code every year. Many others do as well. We do it to push ourselves. Every year, teams like this one come along with stuff that we haven't thought of yet, and advance the whole program. We learn stuff from every team that we play.
The fact that they have so many involved faculty speaks volumes about their interest in the program. Yes, students will do the work in the end, but it puts them in a much stronger position to have so many faculty to draw from.
-- Len
Many years ago that robocup project was actually pretty much what you are alluding to. A bunch of not very functional robots playing a game of: "shove the ball as close to the net as possible and maybe I will score"
It wasn't until 1999 that the actual system was now considered. This is what I believe made Cornell do so well. We focused on the system. How can an improvement in mechanical design change the way to play the game? How can electrical design help out mechanical layout? Its all connected and that's what was missing for a long time. This is where it can be a problem with 6 PhD's in the same field on such a multi disciplinary system.
Look at the system.. It really hurts when you have the best AI but cannot show it due to poor robots/electronics/vision
Ken
Cornell Robocup 2001-2003
http://www.zeevo.com/
They make a cool Bluetooth SoC that includes not only the RF chunks and necessarily Bluetooth hardware, but also an embedded ARM7TDMI processor core and flash, all on the same chip.
The only problem is getting them to talk to you. I'm an EE for a $20 billion/year Fortune 500 that's currently working on a Bluetooth experiment, and the damned rep won't even return an email to myself or one of my fellow engineers. So all I have to play with is a module I "harvested" from another prototype device from another group. Arg!
blue tooth modules for small runs, prototyping here: http://www.flint.co.uk/products/index.php?manufact urer=all&sg=§ion=Communications+and+Wireless&r ange=Bluetooth+Radio+Systems&pID=000000904
" * The ISM is a Bluetooth Class 1 Module designed to plug into a host motherboard for evaluation purposes or small production runs, it can then be transferred into full volume production with the minimum of design requirements avoiding any lengthy design cycles and lowering design costs dramatically
* Bluetooth 1.1 pre-qualified Class 1 - +6dBm
* Fully pre-approved - no Bluetooth or RF knowledge required from integrator
* Data transfer rates up to 200,000 bps
* Range in excess of 100 metres open field
* Stand-alone module with integral TDK ceramic antenna providing excellent interference rejection - ideal for noisy environments
* Very simple integration with existing embedded application
* Hayes AT style command protocol - no Bluetooth knowledge required
* Serial Port Profile & Audio Profile on module
* Serial Port Profile removes need for Bluetooth pairing, alternatively, security features such as PIN and pairing can be added simply by configuration through the S registers
* Bluetooth activity displayed via 2 LEDs
* All I/O lines accessible via S registers using the AT protocol
* AT Command Set - behaves like modem - Bluetooth stack is made intelligent
* 100m range, but low power - achieved with CSR 2nd generation leading BC2 silicon chipset and TDK ceramic high gain antenna
* Fastest implementation - because no Bluetooth knowledge is required, applications are very fast to develop - excellent time to market
* Interoperability - utilities available to integrate with PC's, PDA's and mobile phones
* No microprocessor or extra Flash- fully implemented within virtual machine, full Bluetooth stack embedded in module - no external stack required
"
list price was 129.00