The goal of this robot is to give me a platform that is upgradeable and customizable so that I would not need to build a new robot chassis for each new idea I want to try out. I have been working on this robot off and on for almost a year now.

I want it capable of entering many different types of robot contests, from Trinity to MOBOT and others. It is being designed to do *everything* like a robot really should. My philosophy says a robot designed to be good at only one task is no better than a dishwasher =P

As a side note, I am keeping a journal documenting more of the specific details. Feel free to check it out if you aren't satisfied with the summarized info I put in this instructable:

Step 1: The Mechanical Design

I designed the entire robot using CAD software.

This is an early concept CAD rendering of it that I made:

I used FEA software to design both the robot arm (see image) and the suspension system. The robot arms consist of multiple servos, and one of them has a controllable gripper. I custom machined all parts using a CNC out of delrin.

I made a video documenting on how I made the suspension system. What makes this suspension system new and unique is that its made from just one single easy to manufacture part!

To give good friction to the custom wheels, I rolled up some sticky rubber and bandsaw'ed a long strip of it. It then went right around the wheel (see images).

The final design is the last image.

This is a video of my robot arm testing below. The robot is drivin by two servos.

<p>Great 'ible! Love the final product. Could you go into a bit more detail though? How was the RoboFest or whatever? Looked fun!</p>
I take it you've never washed dishes by hand then. I used to do it every night for the family when I was real young, and my grandpa or grandmother cooked an actual meal pretty much every night. <br><br>Yeah, dishwashers are VERY useful. Heh. Its a cool robot, but this is what most robot designers do and honestly I don't think the solution's ever gonna be throwing more microchips in, not if you want them commonly accessible - that crap is expensive no matter how much they hide the costs by bumping up material and other prices. Like I said, cool bot, but preferably it should do as many things as possible without even bothering to use the microchip, just like the human platform does, automatically and w/o any slow logic processes necessary for things like getting to and from x, getting past obstacle y, etc.
Hi there, I'm 15 and have no prior experience with programming of any kind (I make BEAM robots); could you give me a brief description and list of these programming languages or at least tell me where I could learn them. Any help would be greatly appreciated, thanks.
Hi I don't answer robot questions in email because it can get quite tedious if it isn't a community effort. Try posting your question here, there are several experts that will probably have good advice: http://www.societyofrobots.com/robotforum (try searching the forum first) John (admin) www.societyofrobots.com
Nice one.. I lik it.. please mail me.... how to build this robo. smkumaran.n@gmail.com
Nice robot, like the wheels, don't pay any attention to people telling you your spokes are crooked >_<
the spokes are crooked for a reason not just looks. to sum them up they are pretty much just a one piece suspension system.
Nice Platform, I know what you mean when you say that your tired about making multi platforms as i do alot of the time, If was was going to replacate your project I would need to have it larger but thats not a problem. plus nice programming skills in C, But I like PBasic and VBasic, I guess its just because I really never gave C and C++ the time to learn but I do know some of it and can follow it to convert it to PBasic.
If you want to program robots, you gotta learn C . . .
It's hard out here for a pimp (robot programmer)! First I learned BASIC for picaxe...then went to a little C, then I started taking online classes with MIT and they teach Python! Gah!!! I'm so tired of learning languages! Not to even mention the g-code/mazatrol I learned for work (I program robotic CNC lathes for living :D) Can't we just have a standard people. I vote for C.
I vote PHP (which really is good for things outside of web development as well)
I don't think PHP would work well in a robotics situation. I can and will be proven wrong me thinks. C seems like a good standard to agree on it. Even Arduino uses a variation of C. I really want to get out of Arduino for basic projects but I've just got to take the plunge to learn about Picaxes and then learn from there.
The idea for PHP is kinda a joke (because it will never happen, most people don't even know you can use it for day-to-day programing). I am positive you COULD use it for a robot, but I don't know if it would be as effective as using C. I can't code THAT great in PHP, I have only seen tons of great things done with it. I cannot code at all in C so my opinion on the matter kinda gets nulled anway... but I know it can be done.
Yeah, PHP works great for Web stuff. Without it, well tonnes of stuff wouldn't exist. However it could be argued that in the absence of it's existence, another language would have been created to fill it's void.
Whoops forgot what I wanted to originally add to this instructable. Kudo's to poster for making their own boards with that many io pins. :O :O
Not really the Parrallax chip is a great chip to make robots with. Knowing C is a plus for the other platforms that use a diffrent MCUs, I am learning to program PICs which does use C so I guess I have no choise but to really hit the books. I know that with C you can do a lot more than PBasic I guess thats why alot of people use it. Just call me an old timmer you stay whith what you first learned and to change is very hard but I'll get it. You talked me into it. Thanks
the OOPic is better =p<br/>
Bravo! I love to learn and build one just like Erp. Take a bow Erp! Gil
Very nice, compact PCB. I have a question for you. Currently I use DipTrace for my stuff and not familiar with Eagle3D. Does Eagle3D really let you custom design microcontrollers? Is this done at the VLSI level? Drag and drop?
Eagle is pretty much the industry standard for electronics design. Eagle3D is an add-on to Eagle for 3D visualization. Google up some Eagle tutorials for more info . . .
What I meant is did you use Eagle to design the microcontroller IC itself, or add external ICs/functionality to it? I'd love to design a controller from the ground up (including the architecture/ISA). Currently I'm looking at using an FPGA such as the Spartan to do so, using something like a SoPC builder. If this Eagle allows similar functionality, I'll have to check it out. Also, very nice website. I wish I had the time to do more stuff like that.
Oh no I used the ATmega2560 for this robot. Unless you're insanely rich and don't mind dropping several million on a fab, designing actual mcu IC's is beyond your reach =P<br/><br/>I am not familiar with any commercially available IC design software . . .<br/>
yes! C is very important...
Eagle isn't the industry standard, but it is just about the best electronics design package you can get on a hobbyist budget. Their freeware version lets you design boards up to 4x6 inches with two layers. The student/non-profit version is $130 and can do much larger boards of up to 4 layers. www.cadsoft.de
Could you post a list of all electronic components (servos, cameram sensors etc.)? I really need it!
cool how much did it cost?
This so awesome! You guys are better than any corporation. When is this going to mars?
Good Job! You won!!!
Thank you so much for posting this instructable. I especially enjoyed looking at how you used FEA software.
A local RC hobby shop I often frequent carries an entire line of plastic suspension wheels very similar to the ones in this Instructable. I would think most RC shops would stock them. I believe they are intended for use in model airplanes, perhaps in the landing gear. They come in a wide variety of sizes and weight capacities. I do not remember seeing any quite as heavy duty as the ones in this Instructable look to be, but they would be a great alternative for those of us without the means of manufacturing such complex components. Great engineering and Instructable.
Your robot should be acting in a moving.
I love the suspension wheels!
give this thing an airsoft gun and train, i mean program it to attack people wearing red, and like people wearing blue. strip down a cheapo semi auto one and just connect the switch to one of the outputs
or a paintball gun if it was bigger. parallax has a color senser. i think vex does too. robots play paintball or airsoft!!!!
I especially like the video of it tracking the white line. It looks sorta cute when it moves its head to follow the line. Maybe you should do a Johnny 5 from short circuit...
or wall-e
Wall-E is a very difficult robot to program, at least what 4mem8 and I are doing. Plus 4mem8 is building the J5 as we speek and we are building a Mech to compete in robogames 09
nice. i doubt my robot will be ready by then.
large microcontroller it claims to have. always nice having the robots look up robotics websites with building instructions... invasion must be coming... admirable robot though. 2 questions: why a wheeled robot and not a legged one? and why not create a programming language as well?
Because wheeled is 20x easier than a legged one. =P<br/><br/>Actually, I'm working on a legged robot that will use many ideas I develop on ERP.<br/><br/>No need to create a language, C works fine . . .<br/>
then should i abandon the language i'm creating and just go C?
Which microcontroller dud u use and which programming language? (and by the way nice robot and you website!)
I believe you missed that step ;)<br/><br/>Anyway, I invented my own and its programmed in C:<br/><a rel="nofollow" href="http://www.societyofrobots.com/axon/">http://www.societyofrobots.com/axon/</a><br/><br/>(and now I'm selling them)<br/>
This is the coolest thing I've ever seen!
soo dude thats a lot of money for me $4k is Rs.1,60,000 & i am only 14 so i dont think so my parents are gonna allow me to spend Rs.1,60,000 ... thanks man ....but if i make a robo without using a camera & designing it & only remote controlled den how much will that cost .. & i only wanna make a robo which can move fwd & move its hands pick up things & i dont want it to bump into things .....thanks a lot for helping me man ..bad luck
A servo costs about $10. Two wheels and an arm is 5+ servos. You do the math =P<br/><br/>if you need more help, just go to my forum:<br/><a rel="nofollow" href="http://www.societyofrobots.com/robotforum/">http://www.societyofrobots.com/robotforum/</a><br/>
thanks yaar 4 helping me .........
dude you are really very cool ........... please can u tl me how much will dis cost making that robot & if it is very expensive please tell me some thing which are cheaper ...... i cannot spend more then $20.i will be very thankful to you if you could help me
This has been asked already . . . here's a tutorial for $50:<br/><a rel="nofollow" href="http://www.societyofrobots.com/step_by_step_robot.shtml">http://www.societyofrobots.com/step_by_step_robot.shtml</a><br/><br/>(building a robot for $20 is like buying a car for $100)<br/><br/>quote:<br/>Hmmmm really hard to say actually . . . I probably fried $150 in servos, probably another $400 in servos . . . About ~$2k to design the microcontroller, I got the camera for free but its like $200 normally, about $300 in delrin . . . so it cost me a lot =P<br/><br/>I plan to spend more on it, so it'll probably cost me like $4k in the end.<br/><br/>(its no beginner robot, for sure!)<br/>

About This Instructable


109 favorites


More by societyofrobots: Make a Voice Controlled Robot The Glowing Green Robot Experimental Robot Platform
Add instructable to: