Conference vmszoo::rc

    Quite a project!
    
    Go out and pick up a copy of RC Modeller. In here you want to look for
    a plane called a Senior Telemaster. This plane is currently marketed by
    Hobby Lobby. The plane has been used to fly video cameras and such. You
    will also find various gyros listed. These are generally used to
    stabalize a single axis in helecopters (and some planes). You might be
    able to have your device provide additional inputs to a modified gyro
    for doing things like circling a point/VOR. RCM is well known for being
    a catalog of the industry (full of ads). You can also find various
    altimeters (Casio watch) which will give you density altitude input.

.1  Thanks, I'll look into that
    
I have received several requests to refine my project description.  I will give
the most complete description I can, however please realize that this project
is in the beginning stages and I am sure that changes are going to be made to
just about everything.  I have 8 months to get this project working.

1.  Reason behind the project
I am in computer engineering at RIT.  During Fall quarter of my 5th year, a
partner and myself must demonstrate some realtime device which is controlled by
a computer.  This includes building the device to be controlled, interfacing
all sensors, building the computer, and writing all of the software to control
the computer.  

2. My partner and I have an interest in airplanes and we thought that it would
be fun to try and build a model airplane with an autopilot on it.  We have done
some research on the basic idea and feel that it would be feasible.  We can get
small gyros which are used in model helicopters to sense what the plane's
attitude is(relation to the horizon).  If we keep the pattern very simple, and
turning banks within 10 degrees, we should be able to keep it in the sky.  We
might also need some sort of altimeter to make sure that it did not fly into
the ground.  In order to make sure that the plane stayed in the general area,
we needed some sort of system to allow the plane to know where it was.  We 
thought that 1 or 2 radio beacons would serve this purpose.  

3.  The part we are working on now is the radio beacons and the receivers for
the plan.  We discussed the project with our department head and he liked the
idea, but suggested we start with this part, so we could implement it on a model
truck if the plane and its 3 dimensions proved too challenging.  There are two
ways we considered to implement the beacon.  The first required most of the
circuitry to be aboard the plane, and the second requires a more complicated
beacon.  
If we use some sort of homing circuit, better know as a ADF in the aviation
world,  the beacon would be simple.  All it would have to do is send out a
continuous or pulsed signal for the airplane.  The problem is with the
circuitry which would be required for the plane.  I am not sure all that would
be involved because I have not done enough research on it yet.  The second is
to build what is known at a VOR.  This has a directional signal which rotates
360 degrees and a omnidirectional signal which sends a pulse when the
directional signal gets to the 0 degree point.  This makes calculating the
position of the plane simple, since all you have to do is measure the time it
takes to recieve the omnidirectional signal, to the point where the directional
signal hits.  Most of the circuitry involved here would be in the two receivers
it would take to get the two signals.

4.  Here are some specific questions which have been asked.

  a.  What do I mean by tracking?  (speed, direction, position, ect)

      If we use the VOR setup, tracking would mean following the VOR radial to
      VOR, going past it a few hundred feet, turning around, picking up
      a new radial, and following it back to the VOR, and so forth. We would
      not be concerned with our speed.  Direction would simply be TO the 
      beacon.  It is extremely simple to figure out if you are going towards
      or away from it simply by making a 90 degree turn from a radial.  
      Position would be irrelevant.  All you do is provide a set time after
      passing over the VOR before you turn and seek the VOR again.

  b.  Why is weight an issue? What is the limit?

      Our first project is to get the beacon and corresponding receivers
      operational.  We will then decide what mobile unit it will go on
      and build the computer necessary to control it. If we do end up 
      putting this on a model airplane, we will have to make sure that
      the weight can be carried by it.  What is the limit?  I would say
      a pound or two at most.  We do not want to go over a .60 engine 
      because of costs.  This limits the size and weight carrying ability
      of the plane.  

  c.  Where does this have to work?  Over the entire earth?

      This will only have to work over a area of about a mile.  We would 
      take off, get up to altitude, and land under human control.  The 
      transmitters for the Radio Control airplanes will not go too far
      and we want to keep the plane well within sight at all times.  

  d.  How accurate must the positioning system be.  

      It does not have to be too accurate.  If we can tell where the beacon
      is to within 300 feet we should not have a problem .  The main reason 
      for the beacon is to keep the plane within a reasonable distance
      of the landing site.  A combination of wind conditions and maneuvers
      will most likely tend to move the plane farther away from its starting
      point.  The resolution does not need to be too great either.  If
      I can tell within, say, 10 degrees the direction of the beacon, It
      should be enough to get me to it. Now, if we decide that we are going 
      to implement this system on a RC truck, then the would need much
      higher accuracy and resolution.  The reason for this is our department
      head.  Just being able to keep an airplane in level flight and in
      a general area would be enough to satisfy our realtime project
      instructor.  If we move it onto a truck, however, we will need to show
      that we can control where it is to a reasonable degree.  I know that
      the resolution and accuracy from the VOR setup would be pretty high.  

   e. What kinds of terrain are we dealing with.

      If we use the plane, none.  We will only have to deal with the winds
      and the three dimensional aspect of flying.  The truck would be a 
      different case.  We would choose a flat area for that, so it could
      receive the signal.  

Ok, now that I have described the complete project, I would appreciate your
suggestions.  If you have any more questions, I will be more than happy to
answer them.

Thanks for all your help
nate
       

    The "truck" case is just another version of the automonous vehicle and
    there are lots of them out there for reference. The plane would be
    "neat". Don't rule out larger engines. Gas engines in larger sizes are
    about the same price as a good .60 so that maintains your cost some.
    1/4 and 1/3 scale Cubs are another possibility for planes.
    
    One other thing I forgot to mention was that a local glider guy (Helmut
    Lempke) made a glider "wing leveler" that he flew at the NATs. It was
    electrostatic sensors on each wingtip with some circuitry in the middle
    to differentiate things. Someone else in the file might have more
    information or pointers. This would give you bank angle precise enough
    to keep the wings level.

There use to be a fella hanging around this file who built a lunar excursion
robot. I'm sure he have a valuable thought or two on your project. I think
his name was Willy Smith but don't remember for sure. Anybody else remember?

I ran across him not to long ago in the Electro_hobby Notesfile. You might 
want to check there also.

    see 4.104 for info about the creator of Tycho

    I believe Reith is right about large engines; use an engine from a
    string mower or lawn leaf blower, etc.  (I've seen brand new ones for
    less than $100, and the fuel is cheap.)  But that implies and permits a
    very large plane and hence payload.  Keep the gross weight within the
    AMA limits to avoid insurance problems.  

    Alton, who wants one of the homing circuits to find the VOR station
           that he will bury under the glider landing circle marker.

I've recently been asked about the feasability or doing a "model size" 
autopilot. The plane would be something along the size of a 1/3 scale 
Cub or Ultralight so weight is less of an issue. I just wanted to note 
that the Electrostatic Stabilizers I referred to in .1 were based on the 
work of Maynard Hill and are discussed in the 1986 Symposium Proceedings 
from the Madison Area Radio Control Society (M.A.R.C.S.) National Sailplane 
Symposium.

So, who can answer question on the use of gyros for heading maintainance?

    Gyros come in different flavors.  Rate gyros respond to acceleration in
    the sensitive axis.  These are the commonly available gyros for RC use.
    A rate gyro will not in itself maintain a constant heading because it
    only responds to rate of change of heading.
    
    Charlie
    

The navigation aspect of the project will be handled separate from this. What 
I need is something to smooth out the bumps and correct for the gusts in real 
time. If the tail gets kicked sideways, I want the rudder to compensate and 
put me back in the original orientation to within some small number of 
degrees. If I have a platform that I can consider being stable for some small 
number of seconds, the nav aspect is doable.

    Sounds to me exactly the same as the gyro used in model heli's for tail
    rotor!
    
    Just hook it up between the receiver and the rudder servo, adjust
    sensitivity and BINGO! You have damped yaw control.
    
    Brian
    

There have been several articles in RC Modeler over the last year or so on this
topic. There were by Chuck Cunningham or Don Lowe, and talked about work that had
been done by the author for the DoD. As I recall the articles talked about what
equipment is needed and how it is set up to build the type of autopilot you
are talking about.

Randy

I second Randy's reference to the RCM articles. I seem to recall one of them 
showing a proto type box that gave hands off stability. It might have been in 
the Don Lowe article. Who ever it was made some comments about the need to be
able to turn it on and off remotely so it wouldn't interfere with aerobatics.

On another tack, I read an article on 'fuzzy logic' a couple of months ago.
What had attracted my attention was the picture of the voice controlled RC
helicopter that used fuzzy logic to maintain the helicopter's stability.

 

Now that you mention it, I do remember the RCM articles. I'll have to do some 
digging. I've heard of the fuzzy logic stuff in other applications but not in 
a flight situation. If you can come up with a definate reference or a copy of 
the article, I'd really appreciate it.

I read an article a while back about a company (JR ??) who were producing a 
solid state gyro. The article raved about this, since it has far lower power 
consumption than a conventional one etc. etc.

They demo'd it with an RC bicycle ! 

Since it sounds like you're going to need more than 1 gyro, maybe these will be 
of interest?

Cheers

Nigel

    Yes, it was JR. The bicycle was demonstrated by Graupner at - I think -
    last year's Nuremberg toy fair. People who saw it on TV or in videos
    said it was amazingly slow! 
    
    It is in either this or last year's Graupner news catalog. You might
    want to look them up at Eric's (H.). If you want to get one (or 3 for 3
    axis) and it's not available in the US, let me know.
    
    Best regards,
                   Hartmut