I’m currently trying to design a robot for a scool project and i’d like to have advice.
Here’s some of the constraint.
Energy source is
- 2 x 1.5v AA Battery
- 6 Elastics
Unlimited Potential energy (falling things).
Robot must go, kick two ball into their respective net and come back to original position.
Total value of what is on the robot must be under 100$
No external help to the robot.
Because of the kick ball constraint, precision on position and angular position seam very important to me. Unfortunately time constraint is also an important part of the equation.
However what kill me is the 3v constraint.
Otherwise I’d have got something like the baby orangutan.
I’m thinking about
So I probably need an external micro controller to send drive signal to motor controller.
I also probably need some sort of signal correction if I have the motor and the chip on the same 2AA batteries. (Thinking about one or two large capacitor).
Finally I’m searching for some sort of electric device that can hold the elastic and let it go. I guess it would be in the form of electromagnetically moved mechanical part. (on 3v !?!)
And well there it is.
We are future mechanical engineer.
So electricity is not our best area.
But we can surely understand concepts
This is my attempt to save the idea of robot.
If this fail we’ll probably end up with constant speed DC motor directly on the battery and mechanical hacks for navigation. I’m pretty sure robot-loving peoples here don’t want that.
Those constraints seem pretty difficult. Two 1.5V batteries will get you right to the lower limit of our dual serial motor controllers, so I’m not sure if that will be enough once the batteries aren’t new. In general, the low-voltage motor controller is better suited to the Tamiya gearboxes, so I would recommend that if you have the budget for it.
A better option for 3 V operation is the Low-voltage Orangutan (https://www.pololu.com/products/pololu/0775/), but that costs a bit more and requires a programmer (which probably wouldn’t count in your $100) and requires you to learn C programming for the AVR microcontroller, which isn’t that easy if you don’t have much programming experience.
That 3 V limit really is tough!
Yes for the constraint they are hard on us …
About the 3V input,
Someone told me about a possible step up converter
So it’d go from 3 to 5 or 6 volt for example.
Something like that
rowan.sensation.net.au/elect … tepup.html
what do you think ?
the maximum time for the competition is 120 seconds.
so i do not really care if i drain the battery as long as i can stay online for 120 seconds.
About the programing, you are rigth.
100$ is only about what get on the robot.
Dev kit and programmer don’t count. I sort of knew i had to go for a low level solution. So i tough of doing the strict minimum possible on the chip itself.
For example i’d go with a table dumped on the chip with three columns
-Kick instruction can be a special case of voltage or have it’s own timing column or maybe a sensor …
So the chip only read, wait, and send precalculated values.
Those value could be generated using a language we’re fluent in like matlab.
What do you think ?
Also i we go to the stepper motor direction…
I guess we trash the whole PWM drivers and need to build our own stepper controller ? Existing stepper controller are really expensive.
I’m assuming there are some pretty specific guidelines for like where the starting position is, where the balls are, etc. I can honestly dispute your statement:
This sounds like the sort of thing that a crazy clockwork mechanism would be absolutely great at! I remember a mousetrap car competition in one of my undergrad engineering classes. You just had to drive through two sets of cones, so a circular path worked just fine, but you got extra credit for “active” steering, and the coolest cars were by far the ones with weird music-box like mechanisms. Of course it was a race, so none of these cars technically won. I say if you can’t win, look good while loosing! I’ll see if I can dig up some videos from then.
Also, New Scientist had a neat article a while back called The programmable robot of ancient Greece. Really cool stuff, and maybe some design ideas for you. Unfortunately you need a subscription for the full article text. In a completely unrelated statement, it’s amazing the sorts of things you can find on the internet these days.
Speaking of which, here’s a cool video related to the article. “String programming language,” I love it! Now, you’re probably not allowed to use a falling weight as an energy source, but maybe those elastics are more useful than you think.
Thanks for the encouragements regarding mechanical way :D.
I’d like you to read again the energy source allowed
- 2 x 1.5v AA Battery
- 6 Elastics
- Unlimited (Gravitaional) Potential energy (falling things).
So yea string and mass are allowed
Wow, I read that, then somehow totally forgot it. Yeah, you could really wow everyone and not use the batteries at all. Except maybe as a weight! Seriously though, with those rules, and for such a specific repeatable task, I would probably try to build something completely mechanical. Maybe when that’s done I would use the batteries to power some blinking LEDs.
By the way, I did find a neat video from that mousetrap car competition. This one used pegs on the drive shaft, more of a clockwork or music box mechanism (like the bumps on the rotating drum of a music box that pluck the keys) to knock the steering wheel axel from right to center and back. The rules are you go through one set of cones, then the other, then the first set again, without going outside of the black tape. They had to start at a pretty steep angle, and if you notice the car bumps the first cone ever so slightly (I think you can hear the guy who launched it say “Uh-Oh”). Actually it doesn’t look like they would have made it anyway, but still, very cool.
Anyway, however you end up doing it, let us see, and feel free to keep asking questions (electronic or otherwise)!