geminiz

Hi!

I am currently working on a self automative design
and I have purchased this from Tamiya as the basic platform


I am having trouble figuring out how to automatcially adjust the direction/angle of the car accurately. I am thinking of
using a microcontroller (PIC/AVR/MSP430) and send PWM to a DC motor.
but...not sure how to construct the mechanical part such that the motor
can be used to adjust the direction

edit:
btw I am a mechanical newbie...

I just found out about servo..
after some searching in the forum archive, roughly I understand the electrical concept, but still have no idea
what exactly is its appearance and how it's mechanically linked to the front wheels...



thanks!

__________________
Electrical Engineering Student from University of British Columbia

falleafd

You will not need a servo motor, you should need a servo control for your motor. Because the servo motor turn a fixed angle at a given pulse width. But a servo control give you good controller to maintain the velocity and get the final position. I usually use PID control with commanded velocity and position.

You can search for cheap DC servo control application note from www.microchip.com. It's a very good example. However, to make it simple, you can use P control only.

__________________
Falleaf
mail@falleaf.net
English forums at PIC Vietnam - Vietnamese Electronics forums
R&P Trading and Forwarding Co. Ltd.
Distributor of Microchip in Vietnam

ivancho

You are indeed going to need a servo. Normally most people drive this type of projects with a diferential mechanism, like a tank. Where if you turn your right wheels back and your front wheels forward that would make the tank turn.

But with the car that you have this is not possible, so you are going to need a servo. A servo looks like this


You need a micro-servo, small enoguh to fit up front. The servo moves upt 180 degress. You need to connect the moving mechanism of your front wheels so that when the servo moves it pushes one wheel while it pulls the other one.

Good luck,

Ivancho

geminiz

thanks ivancho (btw ur name sounds so simliar to my Prof's!) and falleafd for the info.

I will give it a try with the micro servo.
hopefully I will be successful on first try and not bother the forum.

__________________
Electrical Engineering Student from University of British Columbia

jbeng

Here's a link for a page which might be helpful to you: http://www.seattlerobotics.org/guide/servos.html
It's a short tutorial on how the servos operate. They're quite easy to work with.

falleafd

Ivan,

I don't think it's a good design if we use the servo motor to control the direction of a 4 wheel car. The fact that, if we solve the kinetic problem of a 4 wheel car, we will find that it needs different directions between the 2 front wheels as the car is turning. And to build the mechanical structure is very difficult with a toy car. That is why I suggest to use 2 active wheels to drive the car with servo control.

How do you think about this?

In my opinion, your design can be used if we don't need the accuracies, and we don't care about the friction from the floor, that is, we need a strong torque of servo motor. But to this application, I think it is oki, and it's easy to use because the toy car is very small.

Try with ivancho's design, geminiz. It's good in this case.

__________________
Falleaf
mail@falleaf.net
English forums at PIC Vietnam - Vietnamese Electronics forums
R&P Trading and Forwarding Co. Ltd.
Distributor of Microchip in Vietnam

ivancho

Remote control cars use a servo to control the front wheel to make a car turn, and sice he already has that base, it is not difficult to implement. Cheap remote control cars use a small DC motor that move some gear to create horizontal movement and pull a wheel out, while pulling the other wheel in.
Kind of like this:



falleafd: I am not sure what you mean The problem with having the servo to turn the front wheels is that the turning radius bacames big. If the wheels where closer and of the same size you could drive it like a tank, and that would allow 360 dregree rotations on the same spot.

Ivancho

falleafd

My English is not good to explain this, however, you should solve the kinetic problem of 4 wheeled car.

The picture you gave is only the damping system, but I'm discussing bout the kinetic problem. That is, the 2 front wheels will never have the same direction when the car is turning.

And as I said, there is no problem with toy cars, And don't care about what I had said.

__________________
Falleaf
mail@falleaf.net
English forums at PIC Vietnam - Vietnamese Electronics forums
R&P Trading and Forwarding Co. Ltd.
Distributor of Microchip in Vietnam

ivancho

Falleaf: I still do't get what you are saying, but I kind of want to

What do you mean: "2 front wheels will never have the same direction when the car is turning"

Ivancho

Nigel Goodwin

Cars never do have parallel front wheels when turning, the inside wheel has to turn further in than the outside one (it's going round a smaller circle). I seem to remember it's set by the 'ackerman angle' in the steering system.

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk

bmcculla

I built an automated car like that back in school. We attached two small bars, one to each wheel steering lever and the other end to one of the small holes in the plastic 'X' in ivancho's picture. The part of the 'X' with the two bars attached stuck strait up when the wheels were strait. It worked fairly well. It didn't compensate for the two slightly different steering angles of the two wheels but for a car thats a pound or two it doesn't matter that much.

Another thing you could try is if you unscrew that plaistic 'X' there is a small gear. You could use that gear in a rack and pinion system.

Brent

falleafd

Yes, thanks Nigel, that is so clear.
Thanks bmcculla, this also what I meant.

__________________
Falleaf
mail@falleaf.net
English forums at PIC Vietnam - Vietnamese Electronics forums
R&P Trading and Forwarding Co. Ltd.
Distributor of Microchip in Vietnam