Mecanical turning system (toy car)

[4Play]

Hello, I need to build a toy car, that will perform the following task:
Go in a staight line for 5 meters, then follow a semi-circunference (or just turn 180 degrees, but not a about-turn) and go back another 5meters, the turning is around a cone:

_____________________________________ ---> boundaries

------------>------->-------\


-------------------cone ---> 0 |


----------< -----------<----/

_____________________________________

I am supplied with a electrical engine, and 3 1.5V batteries, I am not allowed to use any other power source, however I can use pretty much any mechanical or electical device I desire (no remote controls, the car needs to perform on it's own)

I am aware that this is a very simple task, for engineering, but I am in my 1st year and have pretty much only had mathematical classes (calculus, linear algebra, phisics, computing etc) so I am not familiar with the electrical circuits that may accomplish this task...have been brainstorming mechanical solutions for a while now, but nothing really good has come up..

The problem is that the car needs to have 2 operating "modes" : go straight -- turn...and I need to be able to switch between these modes...

Any suggestions would be very welcome :)

p.s: sorry for any misspellings...am from Brazil :p

 

[Danger]

How about:
Have a side-scanning photo emitter/detector or ultrasonic pair to act as 'radar'. When the car detects the cone beside it, it turns for a pre-determined amount of time (say 2-3 seconds; however long it would take to turn 180 degrees) and then goes straight again.

 

[Mech_Engineer]

How about:
Have a side-scanning photo emitter/detector or ultrasonic pair to act as 'radar'. When the car detects the cone beside it, it turns for a pre-determined amount of time (say 2-3 seconds; however long it would take to turn 180 degrees) and then goes straight again.

Or you could use one of those home-made robots that follows a black line drawn on the ground...

I might go for something that has a method of measuring wheel revolutions (and hopefully divisions of a revolution), programmed for how many revs to drive straight, how many to turn, etc.

LINKS:

Line following robot: http://scientificsonline.com/product.asp_Q_pn_E_3053468

Simple programmable robot: http://scientificsonline.com/product.asp_Q_pn_E_3080288

 

[Danger]

That would certainly work as well. I didn't know that you could get stuff like that in so low a price range. Thanks for the links.

 

[titus_polo]

would just like to ask the same question,but were not given any electrical device,its gravity powered car..do u guys have any idea how to make the wheels turn..

 

[Danger]

In this case, all that I can think of would be a series of cams and gears. I'm considering something like how the odometer in a car works. You can drive it from a rear wheel. Calculate how many revolutions of that wheel are necessary to travel from the starting point to the initiation of the turn. The odometer can then trigger a cam and linkage to yard the front wheels over. The second cam (or, simpler, a spring return on the linkage) can then re-centre them after the appropriate distance.

 

[titus_polo]

well,can u think of how many shud i make, for the gears and cams..because the only source of energy would be by dropping 1 kilogram object... and it has to travel at least 20 ft, with 2 turns..btw, tnx for the reply

 

[Danger]

I'm expecting a slow day at work, so I'll fiddle with it there and come up with a basic design with hypothetical figures. You'll have to do the actual calculations yourself, though. We help people here, but we don't do their work for them.

 

[titus_polo]

well the problem is i can't think of a way how to connect the cam with the turning system of the front wheel.
it would be a vertical displacement,ryt?

 

[Danger]

Sorry... I got busy, and again today. Haven't had time to fire up the Mac. I'll do it at the bar after work. (They let me plug in beside the blender, so I can work for hours. )
If you're referring to the steering linkage itself, the displacement will be lateral.

 

[titus_polo]

and another thing to consider,after the second turn,its supposed to move in a straight line,no more turns..can i disengage the cam for that.

 

[Danger]

Exactly. I ran into a buddy at the bar, so it would have been rude to work on the problem there. I'm home now, watching 'Lake House' with W, so I can get on it right away.

 

[titus_polo]

thanks alot.

 

[Danger]

Crap! It turned out that I really liked the movie, so I got nothing done.
Now it's close to 1:00, and I have to get up at 7:30.
Tomorrow is going to be one seriously busy day at work, but I should have enough blank spots to get something figured out. To start with, though, the front wheels should be set up with the same sort of steering mechanism as a real car or RC model, as in a tie-rod system of some sort.

 

[Danger]

I've got it basically figured out, but don't have time to work on it any more tonight. Stay tuned.

 

[Danger]

Started doing it up in Illustrator, but it's too early to link a picture. Suffice it to say that it starts with a worm gear on the rear axle, driving a gear that runs a second axle up to the front. Still working on the cam design.

 

[titus_polo]

well i was trying to make it right now,and haven't try doing the cam,cause really can't figure out how to make it turn for the first 5 feet, then turn it again for the next 4 feet, and about the disengagement,..i just use a simple pulley system to drive the rear axle,just like what u've said,having a hard time luking for gears and stuff..
tnx..

 

[Artman]

How about a string winding system to make the car turn? As the car moves forward it winds the appropriate amount of string from one point to another. Then when the string on one side reaches it's end it yanks the mechanism engaging the steering and slipping the string away. The other side's string is longer and engages later yanking it back to straight and slipping away.

You would need some form of spool to hold the string to be moved, a steering mechanism to hold the end of the string, and a winder activated by the axle motion.

 

[Danger]

I've got it drawn up except for the side view of the front axle assembly. Should be finished and posted today.

 

[Danger]

Okay, finally got 'er done.
Unfortunately, I can't get ImageShack to work.
Anybody else having trouble with it?