Society of Robots - Robot Forum

General Misc => Misc => Topic started by: Steel Neuron on June 20, 2012, 04:36:35 AM

Title: First robot. Some help needed
Post by: Steel Neuron on June 20, 2012, 04:36:35 AM
Hey there!

I'm planning to tackle the $50 robot "challenge". In fact, I'm ready to pay much more than $50, because robotics is my field of study (majoring in electronics engineering), and if I decided to choose this tutorial it is mostly because of the amazing resources and the clarity in which the information is presented. Thank you for that, Admin!

Anyway, the thing is: I am an absolute newbie in terms of mechanical construction, soldering and actually putting together anything electronic. University in my country is really theoretical! But the part I know a bit more about is programming. I have some experience with Assembly, and I've done some assignments at my Uni that involved programming AVR microcontrollers using AVR studio, debugging with a JTAG Ice MkII. I loved doing that and I'd like to fiddle with the code of my robot and try to program it myself.

That said, here come my first problems. I guess I need a debugger, but the JTAG Ice MkII I'm accustomed to using is prohibitively expensive, at least for now. I've seen it goes around $270. So, I was thinking of getting myself an AVR Dragon. What do you think of it? Is it a good tool? Also, I like working on a Linux environment so I wonder if there is a good open source alternative to AVR studio, that doesn't mean shooting myself in the foot too much.

Back to the actual mechanical construction: I wouldn't mind investing in some durable wheels for my robot, considering I'll likely reuse them, so... Is there any alternative for the wheels on the $50 robot (and also the chassis if necessary) that you have used succesfully?

Thank you in advance for your help. If I have any more questions I'll keep posting here!
Title: Re: First robot. Some help needed
Post by: beachboy612 on August 29, 2012, 04:49:13 PM
go to budgetrobotics.com for wheels and servo mounts.  they are reusable.  also, just start it.  dont worry about what you do or do not know.
Title: Re: First robot. Some help needed
Post by: waltr on August 29, 2012, 05:59:07 PM
There are a number of debugging methods that do not require expensive hardware, just some innovation in software. One that I use often is to write values out the processor's serial port (UART) to a terminal program on a PC. Another is turning an output pin on/off that can drive and LED, an O'scope or other device. This can be used to determine If a piece of code ran or to measure timing of a software process.

Break down the debugging to pieces of code then think of ways to test that pieces of code.
Start with these, think up others and you will learn much more about programming and interfacing with hardware.

As beachboy612 said, just start building it. Look through the threads on the $50 robot and see what other people have used for chassis, wheels etc. They look around to see what you have that would work. That really makes it a fun project.
Title: Re: First robot. Some help needed
Post by: knossos on August 29, 2012, 07:52:13 PM
I don't really recommend the "Just start building it" method.  You are likely to spend more money than you would have if you did some research first and asked some educated questions before getting right into it.  For example I purchased some motors for and a motor controller for one of my first projects because I planned to "just start building it" and then found that the motor mounting was a real pain.  Those motors are still just sitting on a shelf because they weren't convenient to use.  I later purchased some motors that had mounting brackets available and have been using them ever since.  Also the motor controller I had originally purchased also just sits there since it was underpowered for the motors I finally settled on.  Another consideration, would you like to add encoders in the future?  If so I would recommend either purchasing motors with encoders now, or motors that have an easy way to add them later.  But all that said, you shouldn't postpone too long, find a balance between doing the research and planning, and actually getting your hands dirty and getting it done.  Above all don't be afraid to fail, that's usually when you learn the most.