R/C car control with serial port
J.J.
I'm trying to make an toy car controlled by serial port.
How can I make a robot, wire-controlled by serial port of a PDA?
I have an old R/C car. It's controlled by radio controller.
The drive train is a 7.2V dc motor, and steering is controled by a
coil, that I think electromagnet.
Direction and speed control is needed for motor.
For coil, direction control is enough.
To communicate, serial port signal process is needed.
There is some off-the-shelf circuit kits in consideration.
http://www.seetron.com/pdf/ssc2_mnl.pdf
http://downloads.solarbotics.com/PDF/KCMD-L298_Compact_motor_driver.pdf
Is it possible to control the speed by the pwm pulse from the ssc2 ?
or Is there any kits suitable for my purpose?
Thanks,
RoboRealm
Have you had a look at http://www.endurance-rc.com/ They may have a
device that allows you to control an RC car from a PC via the trainer
port on your existing RC transmitter.
STeven.
www.roborealm.com
J.J.
STeven,
Thanks for the information.
I don't wanna ues RC control box.
Just contol from the serial port to motor WITH CABLE.
Thanks,
J.J.
Wayne Lundberg
"J.J." <jje...@ouray.cudenver.edu> wrote in message
news:1157572218.3...@d34g2000cwd.googlegroups.com...
your servos otherwise you will have to replace all your RC stuff and use H
bridges and the most troublesome of all, a variable speed DC controller. You
might want to look at using the trainer mode on your RC gear, bypass the
actual RC transmitter and just use your computer to drive the system.
Endurance seems to have the perfect solution.
Wayne
Donald
Hi JJ,
Do you have a picture of the car in question ?
Pics of the insides would be good too.
Interfacing to the motors/servos will need to looked at.
I am sure this can be done, but it sounds like there is no actual "RC
servos" in this unit.
So, a hands on approch will be needed, so I(we) would need to "see" what
you actually have in there.
thanks
donald
Endurance
"Just control from the serial port to motor WITH CABLE. "
Just to be clear, you want to control the car directly from the PDA via
a serial cable? Basically you want a tethered r/c car? If this is what
you want to do, you will probably have an easier time building a car
from scratch. 2 motors could be controlled via a microcontroller to
make a simple car (you can take it further if you wish). It would save
you a ton of time from having to understand what's going on inside
the un-documented r/c car you have and then program the microcontroller
to send the proper signals.
But why not just tap into the transmitter itself? This would allow you
to avoid all the complicated mess of trying to figure out what the
electronics inside the car is actually doing. Here's a link to a
similar project: http://www.hackaday.com/entry/1234000060029935/
The method in the link might be a little different since the car is
only operated by simple button presses for throttle and steering and
not a proportional control system. They also use the parallel port
instead of serial but that would not be too hard to change.
Post a picture though; I would like to see what we are dealing with.
-Endurance
J.J.
I'm sorry to have no space for uploading photo.
It seems that there is no servo mechanism in it.
Just one dc motor and one coil(electro magnet) for steering.
If servo motor, it should have three lines, I think, but it has only
two.
In these days, I visited lots of web pages and found that
there is some difference meaning between RC and RC car.
I thought that RC means that remote control,
but most of RC car means that it has RC servo mechanism.
My car does NOT have RC servo mechanism.
It has simple functions for forward, backword, steering left and right.
My PDA is small and the car is also enough to load the PDA.
To make it easy, I just want to use the motor, steering mechanism and
frames.
There is another choice that using the contol box and loading the
control box on the car. It looks stupid but seems to be easy way, since
there is already an example.
However, another problem rise.
The steering mechanism need time and it does not powered before forward
and backward. Thus, the steering wheel does not move when controlling
speed.
Now, I have another good reason to control the motor and coil directly.
Thanks,
J.J.
J.J.
http://ouray.cudenver.edu/~jjeong/P9080002.JPG
http://ouray.cudenver.edu/~jjeong/P9080008.JPG
Endurance
> It has simple functions for forward, backword, steering left and right.
Knowing this you can find out what voltage drives the motor and the
steering. Since they are just on/off you could program an mcu to throw
some relays.
Donald
Yes, its true, there is no servo.
If you would have said that in the beginning....
Anyway.
Please post a pics of the remote control unit.
By the looks of it, I would say that the steering is controlled by an
H-bridge.
I am sure that the steering is ALL LEFT or ALL RIGHT. Is this true ??
Does the control unit have two buttons, i.e. LEFT / RIGHT ??
OK.
A micro controller with three outputs. One for LEFT, one for RIGHT, one
for 'drive'.
If my assumptions are correct, current through the steering coil will
pull the steering mechanism in one direction or the other.
steering coil
==
----|==|----
==
+ - pulls right
- + pulls left.
The third output will turn on/off the drive motor.
The micro controller should have an internal serial port that will
communicate with the PC.
This looks very simple electric-ly speaking.
The hard part is in writting the code on both sides of the RS-232 ( or
whatever ) you would like to use.
good luck
donald
Clifford Heath
> Now, I made temporary homepage and 2 images.
This car looks very similar to one I converted to MCU control.
The circuit board has three functions: receiver, decoder,
H-bridge drivers (two, each has six transistors). The decoder is
the only IC. It's connected to the H-bridges by four resisters,
one each for forward, back, left and right. If you trace the
H-bridges back from the motor wiring - each of the four wires
is connected to a high-side and a low-side transistor - you'll
find the smaller transistors that enable them - the control
resistors are connected to the bases of these drive transistors.
Measure the voltage levels on the control resistors - mine
switched from about 2V to 5V IIRC - and unsolder them. Build
a four-channel level translator and you can attach four of your
MCU outputs to the drive transistors. Just don't enable forward
and reverse at the same time, or you'll smoke the output.
I also brought out the four wires from the decoder to inputs
on the MCU, so I can use the transmitter to send it control
signals. All eight wires that were hacked into the board go
to a single connector, and I have a "null modem" plug that
reconnects the circuit to function as originally designed.
The next problem you'll encounter is that the main drive motor
probably won't start until you give it about 4 volts, and then
the car will take off like a rocket. I solved that by opening
the axle/motor casing and fitting an optical detector on the
axle. There wasn't much room in there - it's in the conical
end section. The encoder provides velocity feedback (though not
direction) so you can make the motor start smoothly and run
slowly.
Pictures etc at <http://polyplex.org/cjh/electronics/car.html>.
Clifford Heath.
D. Jay Newman
> It seems that there is no servo mechanism in it.
> Just one dc motor and one coil(electro magnet) for steering.
> If servo motor, it should have three lines, I think, but it has only
> two.
R/C Servos are just one example of a servo motor.
> In these days, I visited lots of web pages and found that
> there is some difference meaning between RC and RC car.
> I thought that RC means that remote control,
R/C means either Remote Control or (more typically nowadays)
Radio Control.
> but most of RC car means that it has RC servo mechanism.
> My car does NOT have RC servo mechanism.
Sorry, but R/C doesn't mean actual R/C Servos.
> It has simple functions for forward, backword, steering left and right.
Yes, but they still probably use the same R/C standard as for control of
servos: a pulse of 1ms - 2ms every 50ms or so.
There are probably two channels (one for steering and one for
direction).
> However, another problem rise.
> The steering mechanism need time and it does not powered before forward
> and backward. Thus, the steering wheel does not move when controlling
> speed.
I'd consider getting another car.
> Now, I have another good reason to control the motor and coil directly.
Ayup. Good luck!
--
D. Jay Newman ! Author of:
j...@sprucegrove.com ! _Linux Robotics: Programming Smarter Robots_
http://enerd.ws/robots/ ! "Heros aren't born, they're cornered."
J.J.
Clifford,
you give mea hope to do it.
but i can not find the information of the miniboard.
Thank you.
J.J.
Clifford Heath
> you give mea hope to do it.
> but i can not find the information of the miniboard.
The Miniboard is not the right choice any more - it uses an MC68HC11
which is only UV erasable - no flash. Get an AVR or something...
Unless you are confident you can fit and program an optical encoder
on the back wheels, forget about it - the rear motor is impossible
to control without feedback. Use a model tank instead, their motors
have much taller gearing, and the steering algorithms are easier.
J.J.
Thank you for the information.
I agree that servo motor is easier to control.
Does it be done with only ssc2, if I get a new car with servo motor?
Thanks,
J.J.
J.J.
regulator.
I give up the speed control on circuit level
since my PDA might do PWM control.
This is not a off-the-shelf circuit and it needs some hard work to me.
Do you have any example of this circuit?
I also read a document about the STK500-AVR starter kit,
but I'm not sure I can do that.
Thnaks,
J.J.
D. Jay Newman
Any microcontroller can do this.