Senior project involving tracking an infrared light

K

Ken W

Guest
My son is working on a senior project, that involves a remote controlled car
and an infrared light beam. When the infrared beam is pointed onto a hard
surface (wall, desk, ect...) the remote controlled car will drive itself
toward the beam. He has a prototype, but is unable to get the sensor in the
car to detect the beam at more than 1 or 2 feet. I am looking for someone
that might be able to supply some information or ideas about how to make
this work. For a working project my son will be able to shine the infrared
beam on the wall and the car will drive toward it, basically a laser guided
car. Any help with this would be appreciated.

Ken W.
 
On a sunny day (Mon, 24 May 2004 22:53:00 GMT) it happened "Ken W"
<Havic-man@optonline.net> wrote in
<g1vsc.43368$cz5.14124828@news4.srv.hcvlny.cv.net>:

My son is working on a senior project, that involves a remote controlled car
and an infrared light beam. When the infrared beam is pointed onto a hard
surface (wall, desk, ect...) the remote controlled car will drive itself
toward the beam. He has a prototype, but is unable to get the sensor in the
car to detect the beam at more than 1 or 2 feet. I am looking for someone
that might be able to supply some information or ideas about how to make
this work. For a working project my son will be able to shine the infrared
beam on the wall and the car will drive toward it, basically a laser guided
car. Any help with this would be appreciated.

Ken W.
Is he using modulated light, say on / off at some kHz?
Click to expand...
In such a case it is easier to discrimate between the beam and other light.
JP
 
"Ken W" <Havic-man@optonline.net> wrote in message news:<g1vsc.43368$cz5.14124828@news4.srv.hcvlny.cv.net>...
My son is working on a senior project, that involves a remote controlled car
and an infrared light beam. When the infrared beam is pointed onto a hard
surface (wall, desk, ect...) the remote controlled car will drive itself
toward the beam. He has a prototype, but is unable to get the sensor in the
car to detect the beam at more than 1 or 2 feet. I am looking for someone
that might be able to supply some information or ideas about how to make
this work. For a working project my son will be able to shine the infrared
beam on the wall and the car will drive toward it, basically a laser guided
car. Any help with this would be appreciated.

Ken W.
Click to expand...
I don't know what kind of 'sensor' he is using, but the usual way to
do this type of thing is to use either a Position Sensing Diode (PSD)
or simply a dual diode. For either sensor, the difference between the
photocurrents at the A and B terminals is used to drive the steering.

Sensitivity of the system is controlled by many factors, but the main
design goal is to produce maximum contrast between the source light
and ambient light. Here are some generic ways of doing this:

1) Make source light stronger.
2) Make receiver optics more directional (within the limit of the
required range of control)
3) Use narrowband source (e.g., laser) and make receiver optics
wavelength selective
4) Modulate source and use demodulating receiver electronics (as
pointed out by another repondent)

I'll give some hints on some of the above:

2) Choose an auxiliary lens for the receiver that maximizes energy
collection (large diameter) while minimizing angular coverage (longest
focal length that still provides required angular coverage)

3) Use absorptive or dielectric filters over the detector(s) to reject
most ambient light while passing the source light

4) Pulse the light source. Usually a high intensity and low duty
factor is best (as opposed to a square wave). Discriminate pulse
frequency and measure pulse amplitude using circuits or software at
the receive end. This is simpler than it sounds. Example of type of
processing/circuits: maintain an overall average value and subtract it
from the average peak value for each of the A and B signals. Servo
feedback = processed A minus processed B

More info here:

http://www2.whidbey.net/opto/LEDFAQ/The%20LED%20FAQ%20Pages.html#Links

Paul Mathews
 
This is Ken's son Steven

My prototype uses 6 QSD733 infrared photodarlingtons (Fairchild
Semiconductor) arranged in a 120ş arc around the front of the toy car. The
beam is generated by a infrared LED attached to a 4.5V power supply and a 1k
resistor. When the LED is touching one of the sensors, the sensor assembly
output is 2V despite the 5V bias. At about a foot away the output drops
below 1V and isn't recognized by the comparators I am using. I have a
superbright infrared flashlight on back order from RSR electronics to try
and solve the problem. Any more insight or confirmation on whether or not
the improvements will work is appreciated.

Steven W.



"Paul Mathews" <optoeng@pioneernet.net> wrote in message
news:ed443a10.0405250716.43502d83@posting.google.com...
"Ken W" <Havic-man@optonline.net> wrote in message
news:<g1vsc.43368$cz5.14124828@news4.srv.hcvlny.cv.net>...
My son is working on a senior project, that involves a remote controlled
car
and an infrared light beam. When the infrared beam is pointed onto a
hard
surface (wall, desk, ect...) the remote controlled car will drive itself
toward the beam. He has a prototype, but is unable to get the sensor in
the
car to detect the beam at more than 1 or 2 feet. I am looking for
someone
that might be able to supply some information or ideas about how to make
this work. For a working project my son will be able to shine the
infrared
beam on the wall and the car will drive toward it, basically a laser
guided
car. Any help with this would be appreciated.

Ken W.

I don't know what kind of 'sensor' he is using, but the usual way to
do this type of thing is to use either a Position Sensing Diode (PSD)
or simply a dual diode. For either sensor, the difference between the
photocurrents at the A and B terminals is used to drive the steering.

Sensitivity of the system is controlled by many factors, but the main
design goal is to produce maximum contrast between the source light
and ambient light. Here are some generic ways of doing this:

1) Make source light stronger.
2) Make receiver optics more directional (within the limit of the
required range of control)
3) Use narrowband source (e.g., laser) and make receiver optics
wavelength selective
4) Modulate source and use demodulating receiver electronics (as
pointed out by another repondent)

I'll give some hints on some of the above:

2) Choose an auxiliary lens for the receiver that maximizes energy
collection (large diameter) while minimizing angular coverage (longest
focal length that still provides required angular coverage)

3) Use absorptive or dielectric filters over the detector(s) to reject
most ambient light while passing the source light

4) Pulse the light source. Usually a high intensity and low duty
factor is best (as opposed to a square wave). Discriminate pulse
frequency and measure pulse amplitude using circuits or software at
the receive end. This is simpler than it sounds. Example of type of
processing/circuits: maintain an overall average value and subtract it
from the average peak value for each of the A and B signals. Servo
feedback = processed A minus processed B

More info here:

http://www2.whidbey.net/opto/LEDFAQ/The%20LED%20FAQ%20Pages.html#Links

Paul Mathews
Click to expand...
 
On a sunny day (Wed, 26 May 2004 00:12:02 GMT) it happened "Ken W"
<Havic-man@optonline.net> wrote in
<mhRsc.49358$cz5.18094622@news4.srv.hcvlny.cv.net>:

This is Ken's son Steven

My prototype uses 6 QSD733 infrared photodarlingtons (Fairchild
Semiconductor) arranged in a 120ş arc around the front of the toy car. The
beam is generated by a infrared LED attached to a 4.5V power supply and a 1k
resistor. When the LED is touching one of the sensors, the sensor assembly
output is 2V despite the 5V bias. At about a foot away the output drops
below 1V and isn't recognized by the comparators I am using. I have a
superbright infrared flashlight on back order from RSR electronics to try
and solve the problem. Any more insight or confirmation on whether or not
the improvements will work is appreciated.

Steven W.
You need to pulse the LED, for example with a 555 and transistor.
Click to expand...
Then in the receiver, use an AC amplifier after the photo darlingtons.
With a bandpass filter ...
The filter should pass the frequency of the LED.
Then after the filter an AM (RF amplitude) detector.
This is perhaps the simplest form.
It is a lot more components, and you can get very sophisticated if you want.
(better sensitivity).
Just a start, think about it perhaps.
JP
 
On Mon, 24 May 2004 22:53:00 GMT, "Ken W" <Havic-man@optonline.net>
wrote:

My son is working on a senior project, that involves a remote controlled car
and an infrared light beam. When the infrared beam is pointed onto a hard
surface (wall, desk, ect...) the remote controlled car will drive itself
toward the beam. He has a prototype, but is unable to get the sensor in the
car to detect the beam at more than 1 or 2 feet. I am looking for someone
that might be able to supply some information or ideas about how to make
this work. For a working project my son will be able to shine the infrared
beam on the wall and the car will drive toward it, basically a laser guided
car. Any help with this would be appreciated.

Ken W.
Click to expand...
Isn't your son a bit old for you to be doing his projects for him?
 
Sounds like maybe you're trying to steer the car from the front, i.e.,
you direct the beam toward one side or the other to get it to steer
that direction. If that's the case, you have a chance of getting your
basic approach to work if you can get your beam narrow enough, i.e.,
concentrating the beam power into a small cone. You can use either a
lens or a reflector or both. You might find that a 'Maglite'
flashlight focused to the most intense spot possible works better than
any LED.
Paul Mathews


"Ken W" <Havic-man@optonline.net> wrote in message news:<mhRsc.49358$cz5.18094622@news4.srv.hcvlny.cv.net>...
This is Ken's son Steven

My prototype uses 6 QSD733 infrared photodarlingtons (Fairchild
Semiconductor) arranged in a 120ş arc around the front of the toy car. The
beam is generated by a infrared LED attached to a 4.5V power supply and a 1k
resistor. When the LED is touching one of the sensors, the sensor assembly
output is 2V despite the 5V bias. At about a foot away the output drops
below 1V and isn't recognized by the comparators I am using. I have a
superbright infrared flashlight on back order from RSR electronics to try
and solve the problem. Any more insight or confirmation on whether or not
the improvements will work is appreciated.

Steven W.



"Paul Mathews" <optoeng@pioneernet.net> wrote in message
news:ed443a10.0405250716.43502d83@posting.google.com...
"Ken W" <Havic-man@optonline.net> wrote in message
news:<g1vsc.43368$cz5.14124828@news4.srv.hcvlny.cv.net>...
My son is working on a senior project, that involves a remote controlled
car
and an infrared light beam. When the infrared beam is pointed onto a
hard
surface (wall, desk, ect...) the remote controlled car will drive itself
toward the beam. He has a prototype, but is unable to get the sensor in
the
car to detect the beam at more than 1 or 2 feet. I am looking for
someone
that might be able to supply some information or ideas about how to make
this work. For a working project my son will be able to shine the
infrared
beam on the wall and the car will drive toward it, basically a laser
guided
car. Any help with this would be appreciated.

Ken W.

I don't know what kind of 'sensor' he is using, but the usual way to
do this type of thing is to use either a Position Sensing Diode (PSD)
or simply a dual diode. For either sensor, the difference between the
photocurrents at the A and B terminals is used to drive the steering.

Sensitivity of the system is controlled by many factors, but the main
design goal is to produce maximum contrast between the source light
and ambient light. Here are some generic ways of doing this:

1) Make source light stronger.
2) Make receiver optics more directional (within the limit of the
required range of control)
3) Use narrowband source (e.g., laser) and make receiver optics
wavelength selective
4) Modulate source and use demodulating receiver electronics (as
pointed out by another repondent)

I'll give some hints on some of the above:

2) Choose an auxiliary lens for the receiver that maximizes energy
collection (large diameter) while minimizing angular coverage (longest
focal length that still provides required angular coverage)

3) Use absorptive or dielectric filters over the detector(s) to reject
most ambient light while passing the source light

4) Pulse the light source. Usually a high intensity and low duty
factor is best (as opposed to a square wave). Discriminate pulse
frequency and measure pulse amplitude using circuits or software at
the receive end. This is simpler than it sounds. Example of type of
processing/circuits: maintain an overall average value and subtract it
from the average peak value for each of the A and B signals. Servo
feedback = processed A minus processed B

More info here:

http://www2.whidbey.net/opto/LEDFAQ/The%20LED%20FAQ%20Pages.html#Links

Paul Mathews
Click to expand...
 
Ken- put small plastic lenses over each phototransistor. That is pretty
much free optical gain. Also, use another lens to focus the output of the
LED that projects the IR beam. A small piece of tubing to collimate it is a
good thing.

Cheers!

Sir Charles W. Shults III
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top