View Full Version : wifi for robots

07-24-2009, 08:38 PM
I am doing a BE project where i intend to interface wifi with the micro-controller.
specification: i am using a ATmega 16 controller. I have a programmer ready basically a developmental board. it has a rs 232 interface also.
so i googled for the query. most links refer to something wifi to rs 232 converter. some use network bridges.
But nowhere it is mentioned for which controller or which company is producing it. i found a link in mumbai but same result.
we are ready to spend around 5k on it. there are some bridges for 4k...but do not whether to go along or not....
can anyone help me out with this!!!!

07-25-2009, 12:38 AM
The components required will be hard to get.

Here is another Idea -

Make a wireless Serial Uart on a cheaply available BT dongle. If you can do this not only ull help the community but its greatly needed.

The current ones cost about 50 USD + on avg - about rs 2000 + .

If you can achieve this with a cheap dongle ( Rs 100 BT dongle ) and some micro itll be really useful to a LOT of people. But yes dont forget to make everything OSS ;) .

Now this cheap BT serial port would be something :).

07-25-2009, 06:03 PM
If anyone interested to experiment and make an cheap OSS Bluetooth interface module like Vikas suggest, I like to donate around 2,3k rupees for the project. Please note that you should give the design freely on net for anyone to make.

Contact me if you are interested to spend time on this project.


07-25-2009, 11:56 PM
Well...I'll take you on that Kishore...but NOT at that cost!! :D

07-26-2009, 10:52 AM
ok...ill consider that. and one more thing. our department has a zig-bee assembly with them and we can use it. so i had just checked on nex robotics site, a zigbee terminal for Rs.2812
how about using zig-bee. its not very different from bluetooth is it?

07-26-2009, 06:37 PM
The whole point of me pointing out the BT modem was to cut costs and have it available locally and with easy PC connectivity. a BT dongle costs Rs 100 .

zig-bee does not fit the objective, and a couple ( atleast 2 are needed ) is almost like buying 50-60 of those cheap BT dongles.

07-26-2009, 07:00 PM
ok ok...got the point..
hmm..I am from Goa...so i don't have resources available so cheaply...
can u tell me any good place where i can buy the stuff??
mumbai n pune is fine with me...preferably online trading!!!

07-27-2009, 09:53 PM
Well...I'll take you on that Kishore...but NOT at that cost!! :D


I have no commercial interest in this product. I felt many Robotics hobbyists will benefit by bluetooth wireless system,as it is available in most of the mobile phone's and bluetooth donge also very cheap. My interest was only to sponsor this project.

07-27-2009, 11:01 PM
For those who want to start here is one way to do it:
It can be done on Microchip's latest offering ie. PIC32. It has USB host which is pretty easy to use with the Microchip provided source code. Problem is coding the BT software stack. An easy approach is to use stuff from linux kernel and then port it over to pic32.

I am currently trying to do the same but still have lots of stuff pending.
If you want to start with pic32, you can take a look at bytewerx.blogspot.com

07-28-2009, 08:32 AM
Dont you think it is too much of a hassle?

As it is, a lot of people find it difficult to render a project through to completion, with all of its own problems complexities.......

Now starting on a new Micro, and the problems of BT config, HID, Stack etc etc., will be a total fiasco. You agree yourself that its difficult, so imagine the plight of the beginner!!!!

Fooling around all this software will remove the primary intent of hardware-software bonding along with the interest. This is an expert level undertaking , but I think is too much to go through for the task in hand. It is a good project on its own but may not satisfy the OPs requirement.

What I plan to do is to make things SO easy that there's no fooling around with coding, stack, hid, PC side software or anything else. Just a neat plug n play.....

Good luck!

08-01-2009, 05:52 AM
you cant simply buy a PIC32 and use it, dude, they come in fancy packages. you will know, if you have worked with them. And yes, to learn how that thing work will take more time than the project completion time of the OP. (Assuming he learn PIC32 from scratch)

I do have some questions to the OP -

1) Is there a specific reason to use wifi - is it just to transmit 8 bit data or to switch something?

2) IS the wifi you are talking about same as the wifi that i am talking about - the one that follows 802.11??

3) how do you plan to control the wifi enabled board? - do you plan to use a wifi enabled laptop/computer, if so, which platform do you plan to write the code?
If not, do you plan to use another micro with wifi-converter-something to control the original board that we are talking about?

4) do you expect mega 16 to handle the stack??

Its meaningless to use wifi on anything which doesnt have the purpose of handling huge volumes of data. so if is bytes of data that you transmit, forget wifi.

08-04-2009, 12:20 AM
A standard serial cable replacement via bluetooth module should be a good match for your needs. It is straightforward to use, should hopefully come within your price range of 5k including customs duty, and is wireless. Its just like a plug-n-play replacement for a serial cable.

All you need to do is connect a bluetooth dongle to your computer, install the dongle's drivers which will create a virtual serial port on your computer; connect the BT Modules RxTx pins where you would have otherwise connected your serial cable's RxTx pins and finally, send and receive serial data via the virtual serial port just like regular serial port.

I've used the older model from Sparkfun and it worked excellent. Now they have come out with a new model of the BT Bridge; see this link http://www.sparkfun.com/commerce/product_info.php?products_id=8469


08-04-2009, 01:10 PM
@milindshastri - Idea is to build a module not buy it :) . Build using those Rs 100 BT modules and get everything at very low price. Simulate a USB host on a uC and the BT stack.

We all know that BT to serial are available at 70 $ + usually .

08-04-2009, 02:28 PM
may be the following links helps..
http://instruct1.cit.cornell.edu/courses/ee476/FinalProjects/s2007/blh36_cdl28_dct23/blh36_cdl28_dct23/index.html AVR USb-HOST


@kishore.. I would like to do it..

08-04-2009, 05:24 PM
Aah...in that case having a device like this in our cost sensitive Indian market would be great. I hope you guys make it open source too. Good luck.


08-06-2009, 08:38 PM

the project is about controlling the bot over internet....
so..yah we are making a prototype, so in that case a bluetooth will be good for us....but not for our guide. he wants it to get as close to actual technology. since wi-max finally working, we were thinking if we could have a wifi controlled bot atleast.
these bots have wifi feature.

but how about using zig-bee???
its better then bluetooth and comes close to wifi as well....

08-07-2009, 05:06 AM
you will need wifi bridge. You just cant design your own - for a college project.! unless you buy modules. So the best choice will be a readymade bridge, which might be expensive.

I dont understand when you said zigbee came close to wifi. Zigbee will not meet your requirement. Well, it will - but then you will have to put a comp in between.

so in the reception side, it will be like -

Internet->Modem->Ethernet->computer->zigbee ------> zigbee->robot

but, I Guess your requireent is

Internet>Wifi Modem--------------> wifi Robot

08-07-2009, 08:36 AM
So I guess there is no way for an internet controlled bot. I have searched web for almost a month now, haven't seen anything besides wifi bridges.
So i think it'll be just wireless control.
And please tell me the source of Rs.100 BT dongles and receivers also.

08-07-2009, 02:14 PM
Most project concepts involve a lot of theoretical thoughts with little thought towards their implementation. A project fails due to time/money/resource/patience/consistent-interest constraints. It also fails due to unreasonable demands and wrong guidance.

It is essential to have a good idea of what one wants to do, before embarking on methods of implementation.

It is still unclear to me about what the OP wants to do ( or wanted to do) in the first place, even at the second page of this thread.

Engineering is the science, discipline, art and profession of acquiring and applying technical, scientific and mathematical knowledge to design and implement materials, structures, machines, devices, systems, and processes that safely realize a desired objective or inventions.
One who does ALL this is an Engineer........

Pardon me, all.... but is this a serious project or just a way of thinking big and falling low process???

Because, if it IS serious then something can be done about it.....

There are plenty of ways this Project can be done.

....IF one is serious........ :!: