Need some advice - would this work? Wifi robot idea.

[GalacticNerd]

Hello,

Today I started my final year of highschool. I'm studying industrial ICT. This year we have to finish a device/presentation. What this would be, we get to pick ourselves, as long as it has to do with our curriculum. At the end of the year (June 2012) we have to present it to a few judges who will grade us on it. It well represent 25% of that year's results.
I was thinking about making a wifi/internet controlled car with a network camera on it. Not trying to be overambitious, I've done a bit of research.

How it would work is the following.

We (a team of two) will make use of an RC toy car chassis, and reuse the driving electronics. An arduino will tap into them to control it. It would be hooked up like so:

Motors -> original RC electronics -> arduino -> ethernet shield -> router --wireless connection--> Laptop.
                                              IP camera -> ethernet cable -^

http://arduino.cc/en/Main/ArduinoEthernetShield
http://arduino.cc/en/Main/ArduinoBoardUno

I don't know if anyone can make sense of this, but in my head it does.
I haven't really looked at the programming side of it. (I am familiar with a few programming languages and I don't mind picking a new one, or some libraries up.)
Is there any problem with this configuration? As long as the laptop, the camera, and the arduino have got fixed IP addresses (by MAC address) in the router, I don't see any problem. Would this be doable like this?

I've based myself loosely on this page; http://www.jbprojects.net/projects/wifirobot/
But I think using the arduino/ethernet shield we are able to use an unmodified router.

[Soeren]

Hi,

It's quite doable, but I cannot vouch for your abilities of course.

[GalacticNerd]

Hi,

It's quite doable, but I cannot vouch for your abilities of course.

Thanks, I just want to make sure there's nothing obvious I'm missing that would make this configuration impossible to do what I want it to.
I'll write a (web based) program that streams the images from the IP camera, to a laptop in the (W)LAN network of the router.
It will also have controls (probably joystick but that's irrelevant.) to drive it around, and some bells and whistles with some settings.

Can anyone recommend me a cheap, wired network camera btw? I don't need it to be wireless, as I'll have a router piggybacking on the robot itself.

[klims]

Sounds like a fun project. Only thing I would change is this:

Motors -> original RC electronics -> arduino -> ethernet shield -> router --wireless connection--> Laptop.

to : Motors -> Motor Controller -> arduino -> ethernet shield -> router --wireless connection--> Laptop.

If it doesn't earn you any extra marks it will be way easier to just ditch the (most likely) crappy electronics and replace with something easy to interface to your arduino.
Also, most toy rc cars don't have proportional control of drive or steering. Just on/off control. Look at something hobby grade, maybe even something with electronic speed controller and servo for steering.

I'm guessing your going to want to go slow with this setup. Doing this with hobby stuff is hard because most stuff is built for performance. I would look into something like a rock crawler. They are made for low speed with loads of torque.

Can anyone recommend me a cheap, wired network camera btw? I don't need it to be wireless, as I'll have a router piggybacking on the robot itself.

Check out dealextreme. I've seen a few reasonably priced units that even have pan/tilt functionality. Make sure you look for devices with good reviews.

Not sure at your level of knowledge, but I've done a few projects like this for uni so hopefully you can learn from my mistake: From what you have posted it sounds like the majority of time on this project will go to making a nice web interface to talk to your hardware, so don't waste the next 6 months messing around with the toys then rush the interface. *Most* teachers don't care if you designed and built the electronics from scratch, they just want to see something that looks cool and works. In my experience at least 20% of the other group will completely fail at getting anything to work because they overextended. Have clear goals you want to achieve, then once you have nailed those go for the icing on the cake.

Good luck!

[GalacticNerd]

Sounds like a fun project. Only thing I would change is this:

Motors -> original RC electronics -> arduino -> ethernet shield -> router --wireless connection--> Laptop.

to : Motors -> Motor Controller -> arduino -> ethernet shield -> router --wireless connection--> Laptop.

If it doesn't earn you any extra marks it will be way easier to just ditch the (most likely) crappy electronics and replace with something easy to interface to your arduino.
Also, most toy rc cars don't have proportional control of drive or steering. Just on/off control. Look at something hobby grade, maybe even something with electronic speed controller and servo for steering.

I'm guessing your going to want to go slow with this setup. Doing this with hobby stuff is hard because most stuff is built for performance. I would look into something like a rock crawler. They are made for low speed with loads of torque.

Can anyone recommend me a cheap, wired network camera btw? I don't need it to be wireless, as I'll have a router piggybacking on the robot itself.

Check out dealextreme. I've seen a few reasonably priced units that even have pan/tilt functionality. Make sure you look for devices with good reviews.

Not sure at your level of knowledge, but I've done a few projects like this for uni so hopefully you can learn from my mistake: From what you have posted it sounds like the majority of time on this project will go to making a nice web interface to talk to your hardware, so don't waste the next 6 months messing around with the toys then rush the interface. *Most* teachers don't care if you designed and built the electronics from scratch, they just want to see something that looks cool and works. In my experience at least 20% of the other group will completely fail at getting anything to work because they overextended. Have clear goals you want to achieve, then once you have nailed those go for the icing on the cake.

Good luck!

sorry for the late reply, check this thread for my progress.
http://www.societyofrobots.com/robotforum/index.php?topic=14396.0
I've gone ahead with using an RC car. you're right about the steering etc, but I'll see how good it works, I can always hack a servo in there if it doesn't.

[a903nx]

Hello, I want to share about my WIFI robot project that I was made a years ago.

For the blok diagram :

   Robot side

   Wireless          Wizznet
   Router <---->  Serial to Ethernet gateway ----> ATmega32 ----> L298 motor Driver
       ^
        |
        |
IP Camera


   Controler Side:

   Notebook with controller software made by me.


I do some image processing to detect moving object with my IP Camera. Wireless router was the communication device to connect Robot and station. The data in IP protocol from wireless router was decoded to serial data by Wizznet serial to gateway router. ATmega32 will control the robot moving.

that's all. maybe its can help you

[GalacticNerd]

I have about the same layout, only I have an Arduino ethernet that has the wiznet chip too.
I haven't planned on any image processing as of yet. I figured I could make an automatic mode with a sharp IR or two.