Friendly competition?

[blackbeard]

Ok so i was just wondering what everyone including admin thought of this. I was thinking that it would be really cool if we had a prizeless competition every once in a while between members where we do something related to robotics. The members can judge for themselves in a pole to decide who's project is better and after that all the projects can be put into a HowTo guide. What this would do is create a pool of projects that all do the same thing but with ideas from all of the participants which would help show people several ways of doing one thing. Also since everyone has their strengths and weaknesses we can take bits and pieces of ideas out of each project to make something better then all of them. It could also encourage new people to try a variety of methods to accomplish a single task rather then following a simple and well laid out guide.

[futurrobotech]

I've seen stuff like this on various programming websites. Where each week there's a new challenge and it's basically intended to get people to use stuff they wouldn't normally use. So perhaps We could do a competition every three months with a challenge.. Say use a PIC to do something then next time use an Atmel, so ultimately we'd all gain even greater exposure. Just a thought

[Joker94]

I think a competition every few months is a good idea.

One thing that i would say that i don't think that we could make it MCU specific as that would reduce the number of people who take part from competition to competition as many people use one or the other for there own reasons and would not be willing to venture into the world of another MCU.

But having regular competitions with a common goal amongst the community would defiantly advantages to many people.

[futurrobotech]

I think a competition every few months is a good idea.

One thing that i would say that i don't think that we could make it MCU specific as that would reduce the number of people who take part from competition to competition as many people use one or the other for there own reasons and would not be willing to venture into the world of another MCU.

But having regular competitions with a common goal amongst the community would defiantly advantages to many people.

It would be good for learning purposes though

[Joker94]

It would be good for learning purposes though

that is a good point

we'll have to wait and see what other members think as the idea progresses.

[blackbeard]

it would allow learning for sure and programmers for pics and atmels are cheep so the cost wouldn't be that high. Languages could also be a good. My first idea was for one of the contests to be about making the cheapest robot platform that performs comparably or better then the $50 robot  essentially to see how low we could go or a contest about what people can make without high end tools (i.e. an oscilloscope, professional multimeter, stationary power supplies and signal generators.) which most newbies don't really have access to.

[Ro-Bot-X]

Years ago I was in a club that had the same idea. Every 3 months we did a challenge, so we would meet every month and talk about what we would like to do and iron out the rules. Then we built the robots and had the challenge with so much fun. But we were using Lego Mindstorms, which allowed us to build different robots every time. I would really like to see something like this done with hobby robots.

Example of challenges:
- Line Following
- Line Maze
- Maze (search and rescue)
- Maxwell's Deamons (sort black and white cubes)
- Temple of Atlantis (build a gate like structure out of cubes and beams)
- Fire Fighting
- Robo Butler (bring food to the table, clean the room, feed the pet, etc.)
- Trash Collection (follow the line, find trash cans and collect the trash, dump the trash at the end)
- many more that don't occur to me at this early morning time

I thing a basic robot chassis can be used for all challenges, plus some additional modules (like a grabbing arm). Everyone will do the challenge at home, document and video it. I think some basic rules like robot max size, etc. can be set, as if the people would compete in a real challenge.

Both mechanical and programming techniques can be improved if people participate or even if only study the docs and videos of other people's robots. So yes, I think this is a great idea!

[Asellith]

The problem with robotics is its expensive to experiment. A programmer can use the same equipment to experiment in hundreds of languages and methods but robotics requires more cash.

Something that might work would be to take a section of Webbot's library for instance and write tutorials and documentation for that section. Then Webbot gets better more indepth examples/documentation and the tutorial count around here goes up.

Also something I've seen is reviews or articles about subjects. For instance we could say write an article based on camera technology. It could be a review of a camera you have or a tutorial for a project using a camera. If you have lots of camera experience you could write an article comparing several cameras. Then we vote on the best. Codeproject is an example of this system. If we do good enough work and someone does the leg work we could start offering prizes that are donated by companies.

[futurrobotech]

Yeah.. I understand that... Money would be an issue.. or we could come up with a list of things we can use like certain sensors and what not and just see what people can make out of it interesting..  I'm not sure.. kind of new to the whole robotics thing and more of a programmer and electronics guy.. I would think we would need Admins opinion before moving forward with anything. I would for him to feel as if we are encroaching on his territory as far as competition wise.. But I would like to have an opportunity to do this because I feel it would put a little bit more of a push behind me.

[z.s.tar.gz]

Or what if the catch is "Build X using the lowest budget"
This leads to low costs and good engineering to make up for the lack of parts.

I for one am in favor of a free-for-all robot buildoff.

[Ro-Bot-X]

Lego Mindstorms is perfect for such competitions, because the kit has a microcontroller, motors and sensors, so anyone that wants to participate starts with the same base kit. Also, by Lego nature, people can use the parts to build a huge number of different robots without the need for a shop to make parts. Build a robot, document, video it, take it apart, build a different one. Easy.

How can we do that with regular parts? Well, any robot needs a microcontroller, some motors and some sensors. We can specify the number for these in the rules. But I think the hardest is to build the structure. Easy to work with materials like plywood and acrylic are good choices and also any Legos you might have or Erector (Mecano) pieces can be used. We have to let the participants be creative.

For instance, a universal electronics set for such a robot can be bought for about $100:
- 2 sets of motor + wheel, $7.98 each: http://www.solarbotics.com/products/gmpw_deal/
- ball caster set, $2.49: http://www.solarbotics.com/products/23200/
- Ardweeny microcontroller, $9.95: http://www.solarbotics.com/products/kardw/
- FTDI breakout board, $14.95: http://www.solarbotics.com/products/50512/
- breadboard, $4.45: http://www.solarbotics.com/products/21025/
- breadboard wires $4.95: http://www.solarbotics.com/products/21010/
- breadboard voltage regulator $5.95: http://www.solarbotics.com/products/34020/
- motor driver H-bridge, $1.75: http://www.junun.org/MarkIII/Info.jsp?item=60
- 3x line sensors, $1.35 each: http://www.solarbotics.com/products/qrd1114/
- Sharp distance sensor, $8.75: http://www.junun.org/MarkIII/Info.jsp?item=9
- Sharp sensor cable, $1.75 each: http://www.junun.org/MarkIII/Info.jsp?item=38
- 2x standard servo, for arms, $5.95: http://www.hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=3743
- 2x micro servo, for pan/tilt, $5.95: http://www.hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=8499

There are a few other parts needed, as a battery (or a AA battery box and 4 AAs), connectors, pin headers... so the price will go up to about $120. Plus various shipments. Of course, there are other options  too, for instance one can use a uBotino and a Prolific USB-serial for $35 and ditch the Ardweeny and the breadboard stuff.

I was thinking to offer a basic robot kit on my store, but the shipping prices to/from Canada are too expensive. To be worth it for me, I would have to charge a little bit more on the parts, so people are better off by getting the motors and sensors from other stores. Sorry about that.

[Soeren]

Hi,

Lego Mindstorms is perfect for such competitions, because the kit has a microcontroller, [...]

For all the AVR only people, perhaps mentioning that it uses a PIC microcontroller would be fair.

And the price is rather variable based on geography - worst deal is here in Denmark (where it's produced) - The LEGO Nxt is more than twice the US price (and the US web site hides the prices when you surf in from DK - had to use a couple of anonymizers to confuse them).
Even our "neighbors" like Great Britain and Germany gets way lower prices - I think LEGO is P'ing big time on their home base.

[blackbeard]

Cost shouldn't always be the deciding factor. Robots are expensive but this IS A robot forum! Since it's not for prizes we don't need rigid timeframes and we can get users opinions prior to starting a contest so we can gauge interest.

[madsci1016]

For all the AVR only people, perhaps mentioning that it uses a PIC microcontroller would be fair.

It's not, because it provides the end user zero real PIC experience or knowledge gained any more then if it were AVR or something else powered. No dealing with PIC toolchains, registers, etc. It's all a blackbox to the end-user. So who cares?

If anything Mindstorms shows of the power of programming using Labview, whether or not you feel that's a good thing.

[Soeren]

Hi,

It's not, because it provides the end user zero real PIC experience or knowledge gained any more then if it were AVR or something else powered. No dealing with PIC toolchains, registers, etc. It's all a blackbox to the end-user. So who cares?

Everyone that wants to hack it to give it some real power by using the controller properly... Plenty of LEGO hackers out there.

[Ro-Bot-X]

From this Wiki page:

The first generation of Lego Mindstorms was built around a brick known as the RCX. It contains a 8-bit Renesas H8/300 microcontroller as its internal CPU.

And from this page:

The NXT actually has 3 microcontrollers making everything tick along.

(1) The little circuit board at the top of the picture is the Bluetooth to serial module. The left most, rectangular chip is the memory with the Bluetooth control software from Cambridge Silicon Radio. The square chip is the CSR Bluecore 4 microcontroller and radio hardware BC417143BQN, supporting Bluetooth v2.0 and EDR. The gold zig-zag trace at the top right hand corner is presumably the antenna.

(2) The main 32-bit ARM controller that runs your programs, an Atmel AT91SAM7S256. This includes Flash memory/file system, RAM and the USB Device interface.

(3) An 8-bit Atmel ATMEGA48. I presume this manages the pulse width modulation (PWM) of the motors, and uses the feedback from the NXT motor rotation sensors to tune the power to each motor.

Where is the PIC microcontroller?

[Soeren]

Hi,

Where is the PIC microcontroller?

Oh, my bad

It was the VEX Robotics controller I confused it with (I don't own any of either toys).
Sorry for the mixup!

[blackbeard]

ok so i pm'd admin but he hasn't responded so i'll post here. Here is my idea and i am wondering what you all think. here goes.

So when i started robotics sensors were always a big barrier for me. understanding and interpreting the data was hard for me and there isn't a superb amount of info that's really hands on so here is my idea. My idea is that the first competition be based on making electronic musical instruments. the goal of this would be to build a device that interprets data from it's sensors and converts it into sound. it could be anything from theramin style with sonar or light sensors to pianos using resistive touch screens and anything else one can think of. this can also help newbies to the electronics world as it can be done without the use of a microcontroller by a clever inventor. it will also help people learn to interpret the data that sensors give us in order to incorporate it  into their projects better.