6843brooksj@vms.csd.mu.edu (06/09/91)
Has anyone on here experimented with hovering robots? I am considering trying to construct something that use three or four small ducted aimable electric fans. Anyone ever try something like this? My main concern is obviously getting fans small/light enough with thrust sufficent to lift the platform. I have experiemented with a small ducted fan that could lift its own weight (and a little more) but I need something quite a bit more powerful. Any help would be greatly appreciated Jack Brooks
chuck@brain.UUCP (Chuck Shotton) (06/09/91)
In article <00949D78.A594B600@vms.csd.mu.edu>, 6843brooksj@vms.csd.mu.edu writes: > Has anyone on here experimented with hovering robots? I am considering > trying to construct something that use three or four small ducted aimable > electric fans. Anyone ever try something like this? My main concern > is obviously getting fans small/light enough with thrust sufficent to > lift the platform. I have experiemented with a small ducted fan that > could lift its own weight (and a little more) but I need something > quite a bit more powerful. Any help would be greatly appreciated > Jack Brooks > How about sticking some smarts in a remote controlled model helicopter? That's proven, cheap technology, and controlling it wouldn't be to hard if you toss in a couple of gyros (poor man's IMU?). ----------------------------------------------------------------------- Chuck Shotton Internet: cshotton@girch1.med.uth.tmc.edu UUCP: ...!buster!brain!chuck "Your silly quote here." AppleLink: D1683 MacNet: shotton
ear@wpi.WPI.EDU (Eric A Rasmussen) (06/10/91)
If you don't mind making the system A) bigger and B) louder, you could try using simple lawn mower engines. I remember hearing that they have sufficient power/weight ratios to lift themselves and a bit of control equipment so that if you put three in a triangle layout you will get yourself a small hovercraft. Of course, this would probably have to be used outside, but that could just make it more interesting. +---------< Eric A. Rasmussen - Mr. Neat-O (tm) >---------+ +< Email Address >+ | A real engineer never reads the instructions first. | | ear@wpi.wpi.edu | | (They figure out how it works by playing with it.) | | ear%wpi@wpi.edu | +---------------------------------------------------------+ +-----------------+ ((( In Stereo Where Available )))
jack@rml.UUCP (jack hagerty) (06/10/91)
In article <00949D78.A594B600@vms.csd.mu.edu> 6843brooksj@vms.csd.mu.edu writes: >Has anyone on here experimented with hovering robots? I am considering >trying to construct something that use three or four small ducted aimable >electric fans. Anyone ever try something like this? My main concern >is obviously getting fans small/light enough with thrust sufficent to >lift the platform. I have experiemented with a small ducted fan that >could lift its own weight (and a little more) but I need something >quite a bit more powerful. Any help would be greatly appreciated > Jack Brooks Many years ago (around 1984, I think) the magazine _Robotics Age_, before it became _Robotics Engineering_, did a short story on a hovering robot. The experimenter was a bit more pragmatic and used a gasbag for lift and thus could concentrate on a control system to simply steer. Went into the problems of scaling a gasbag too (i.e. you loose useful lift as you scale them down. A 1/2 scale balloon has less than 1/2 the lift). I could look this up if you want. Perhaps you could use the gasbag as a "boost" to reduce the power required of your verticle thrusters but still retain the ease of verticle movement without inflating/deflating the bag. - Jack ============================================================================= ||Jack Hagerty, Robotic Midwives, Ltd. jack@rml.UUCP (smart mailers)|| ||Livermore, CA ...!uunet!lll-winken!rml!jack (dumb mailers)|| ||(415) 455-1143 jack%rml@lll-winken.llnl.gov (desperate mailers)|| ||-------------------------------------------------------------------------|| || "The Biblical God is a sloppy manufacturer. He's not good at design, || || He's not good at execution. He'd be out of business if there was || || any competition." - Carl Sagan, _Contact_ || =============================================================================
hollombe@ttidca.TTI.COM (The Polymath) (06/11/91)
In article <00949D78.A594B600@vms.csd.mu.edu> 6843brooksj@vms.csd.mu.edu writes: }... I am considering }trying to construct something that use three or four small ducted aimable }electric fans. ... My main concern }is obviously getting fans small/light enough with thrust sufficent to }lift the platform. I have experiemented with a small ducted fan that }could lift its own weight (and a little more) but I need something }quite a bit more powerful. ... How high off the ground to you want it to hover? If the idea is simply to eliminate wheels and tracks, consider a hovercraft design. Putting a skirt around the base of the platform greatly increases ground effect and lifting ability. I know toy hovercraft are available off the shelf. I don't recall if any are electric powered. If you're controlling fan thrust with the ducts, then a small gas engine may be the power source you need. -- The Polymath (aka: Jerry Hollombe, M.A., CDP, aka: hollombe@ttidca.tti.com) Head Robot Wrangler at Citicorp Illegitimis non 3100 Ocean Park Blvd. (213) 450-9111, x2483 Carborundum Santa Monica, CA 90405 {rutgers|pyramid|philabs|psivax}!ttidca!hollombe
mwtilden@watmath.waterloo.edu (Mark W. Tilden) (06/15/91)
The following are some articles I've kept on the subject. Might be still worth a read. ------------------------------------- To: outram@cernvax.cern.ch Subject: Re: model flying saucer - help/advice needed. Newsgroups: sci.electronics In-Reply-To: <1746@cernvax.UUCP> Organization: University of Waterloo Cc: Bcc: In article <1746@cernvax.UUCP> you write: > ___ > TOP VIEW : / O \ where 'O' is an engine. > |O + O| > \ O / > --- Limit it to three motors and install a microsequencer (rom based) to do high speed dynamic stabilization control for you. Nothing else is fast or light enough. Three motors gives you better lift area while simplifying the control problem to tripod balancing. The actual balance mechanism should be two loops of plastic tubing half filled with murcury (the other half with heavy oil). Paint the surface of the tubes half way with high conductive paint. Stick an insulated wire in contact with the murcury. Feed both outputs into a capacitance meter circuit and you have a damped stabilizer of reasonable accuracy. Far cheaper than building your own gyroscope. >Has anyone made one of these things before ? What difficulties where >encountered ? Yes, I made one out of carved styrofoam as a frame. Biggest problem is finding the ideal lifting fans. As for control, drive all fans from a single high-power electric motor (used in rc cars) driving triple magnetic clutches. You'll have to make these. Hope you have a Unimat lathe handy. The motor, as the batteries, must be mounted dead center. Using 100mA batteries, you can expect a maximum 2 min flight per charge. Longer if you want to run it by 'wire' of course. You will also want a speed regulator for the motor or control instability will result. Your biggest problem is weight and control. Too much weight and all you'll have is a floppy hovercraft. Control is also the major bitch. You require very high speed/accurate feedback to keep the thing from flopping hard. One of the things I did was install a deep murcury switch which only registered if the machine flopped > than 90 degrees. When triggered it replaced the balance circuit control with signals to level the craft at all costs. You won't have to slamm the thing into the ground too many times before you realize how handy this is. I'll tell you now, such a thing really isn't as fun as you might think. It is very difficult to control and if your motors aren't precessed exactly the damn thing is always spinning on you. This won't matter if you can find a gyroscope you can lift but otherwise you'll have hard times ahead. The only way around it is to leave the lift motors always stabilizing the outfit and install control vanes around the outside of the craft which allow retro-rocket like control. Works, but very boring, noisy and unstable. Anyway, that's my two cents. Good luck with it. -------------------------------- Newsgroups: rec.models.rc Subject: Re: Flying saucer - the beast emerges ! Summary: Expires: References: <1765@cernvax.UUCP> Sender: Followup-To: Distribution: Organization: University of Waterloo Keywords: In article <1765@cernvax.UUCP> outram@cernvax.cern.ch (Nicholas Outram) writes: > The craft will be powered by FOUR rc variable speed engines to >provide lift and small amounts of side thrust for horizontal mov- >ment. Two more small electric motors provide spin stability and >rotation capability. I suspect your craft will be far too heavy. Might want to reconsider three lift motors for better lift vs. effective area. As you're using a processor, fast calculation of the centroid of the device will be no prob. > Horizontal stability controlled by Mark Tildens 'mercury gyro' >idea. ... >-> I still need help of how to sense the rotation of the craft <- >-> Something like an electronic compass ?????? <- Just now I've had a better idea which might kill two birds with one stone: rotation and stability. Place 4 double contact mercury switches on a shallow bowl, the hub connected to a small stepper motor. The outside edge of each switch raised to a yet-undetetermined- angle. Rotate at a fixed speed (defined by uP) until the all switches *just* register closed at the outside edge. Any tilt will register as a open in the direction of lift and any rotation will be detected as uniform open contacts compared to the ideal spin motor value (this value being predetermined by experiment). Connections to the switches made by pickups on the outside of the dish. Granted rotation detection can only be made in the counter-stepper-spin direction but if the craft is designed to spin against this direction anyway, then the problem is solved. Granted also the signals coming back might require their own processor for correct filtering and control, but it's a neat idea. The range of values of rotation speed and switch angle should make up for the inherant errors generated by the switches. Wonder if it's practicable? >-> Also how do I convert small capacitance values to a form <- >-> easily input to a computer ????? (0-1nF)^ <- Make it part of a temperature-insensitive transistor based oscillator with a square wave output. Feed this signal into the strobe of a 8-bit latch (positive edge) and a delay line which will reset the 8 bit binary counter which is feeding the latch (at positive edge + a tad). The clock source for the counter must be a Xtal source timed to be approx 128 times faster than your transistor oscillator. Thus, if your t-osc frequency goes up or down, you have a 128 bit window to play in. A simple extension to this circuit will allow you to multiplex multiple capacitive inputs to the same output latch. As your lower 2 or 3 bits of your data will be random error, replace these with a binary value corrosponding to which capacitive tilt meter you're referencing. These will, of course, be driven by their own counter, the clock input from the t-osc. A five chip solution. Is all. ------------------------------------ Re: Flying saucers: >In one of yor artciles you made said that you have been able to construct >such a beast. Could you be a little more detail about these magnetic >clutches? You see, I'd like to make a platform with three rotors, and >I need some way to vary the thrust that each puts out for steering control. >My instaneous idea was to use three motors, and vary the speeds, but that >would be far too heavy. Second idea, use one central motor, driving var. >pitch rotors (mechanical nightmare and heavy). It would seem that >using one motor with three "clucthes" would be ideal, so any ideas in this >area would be greatly appreciated. Thanks, Jack Brooks >6843BrooksJ@vms.csd.mu.edu Yes, that's pretty close, I think. Magnetic clutches are available from a variety of industrial suppliers but the best thing is to find one at a surplus shop. They are small, rotary bearing shaped devices with two wires coming out of them. A good substitute is to find a source of pancake walkman motors and direct drive the shaft. If you short the leads from these motors, the back-generator effect causes a drag proportional to the efficiency of the motor. Pulsing this effect with a single transistor does a good job and is very energy efficient. Fast too. Only problem is the additional weight of the motors. My original flying creature was a flat thing made out of styrafoam. When all was said and done it was not really very exciting and took up far too much power. If I were to do another one, it would look more like this (side view): Magclutch and paddle - #------= Fan vanes and mount - = ----+---- =====|===== ----+---- = no-slip band drive - |----------|----------| drop frame and control- \_ | _/ \|/ Single hi-ef motor - [=] A single, low high-efficiency motor drives three lifting vanes by a no-slip band drive (nylon O-rings). The same shaft which drives the vanes also drives a free-rotating magclutch which is controlled by the onboard intelligence. I found that only a very small reduction in the available airflow to one lifter vane was sufficient to steer the device without damaging it's lifting ability. Saves on control vanes too. By mounting a small 'paddle' on the magclutch, you could limit airflow to any of the three lifters by a simple feedback control circuit. Also, if the paddle position can be controlled from side to side of a single vane tube, you can also control roll and direction. Unlike my original device, I now know that if the majority of the weight is 30 degrees below the lifting points, the device is self-stabilizing, so that helps. Neat thing about this design is that with the new 17% efficient solar-cells available, it could be entirely self-supporting. Could be. I'm sorry, are you building a flyer, hovercraft, model or full-size version? This idea is really only good for a model of some sort. Is all. Mark Tilden. -- Mark Tilden: _-_-_-__--__--_ /(glitch!) M.F.C.F Hardware Design Lab. -_-___ | \ /\/ U of Waterloo. Ont. Can, N2L-3G1 |__-_-_-| \/ (519) - 885 - 1211 ext.2454, "MY OPINIONS, YOU HEAR!? MINE! MINE! MINE! MINE! MINE! AH HAHAHAHAHAHAHAHAHA!!"
bkoball@well.sf.ca.us (Bruce R Koball) (06/19/91)
Re: small RC hovering models...I saw an amazing example of same at a conference last fall...fellow had purchased it in Japan...consisted of 4 small PM/DC motors (the type you see in cheap toys) mounted on a styrofoam ring about 12" in dia.... each motor had a plastic prop about 4' in dia....control electronics and batteries (about 8 oz. worth) in the center...stability was accomplished via a small *GYROSCOPE* (no kidding!) in the center pod...gyro had small, dual-gimbeled flywheel with some sort of proportional sensor feedback (don't recall the method)...gyro parts (except the flywheel and bearings) were injection-molded plastic....owner said the whole thing cost about $400 US....I might be able to track him down if anyone's desperately interested....reply by email Bruce R. Koball Motion West (voice only) 415 540-7503 bkoball@well.sf.ca.us 2210 Sixth St (ans machine) 415 548-2450 bkoball@cup.portal.com Berkeley, CA 94710 (fax) 415 845-3946
alan@km4ba.uucp (Alan Barrow) (06/19/91)
I did some thrust testing of various electric motors used in RC cars
and planes. Most were pretty good Re: power to weight.
The big problem I ran into was battery power to weight. The limiting
factor seemed to be there. Adding or deleting motors did not make that much
difference in the thrust budget, as they were offset by the batteries
required by the config.
My base design goal was a creature that would:
hover itself for 5+ minutes
have enough power surplus to manuver during that period
pickup and carry some small objects.
Based on thrust measurements & available battery data, I could not
get an (electric) critter to hover for 5 min, much less manuver & carry
things. 1-2 minutes looked possible.
So that left the following options-
gas bags for lift- (too clumsy & slow)
dynamic lift- (Not able to hover to pick up objects)
gas engines- (feasable using the mag clutch methods
discussed earlier.)
As the gas engines got into control issues I was not interested
in, I did not persue further. (rpm syncing 3-4 gas engines???)
A single engine approach using a gas engine would make it simpler.
Could you "PWM" the power to the fans using the clutches to vary thrust???
Hmmmmmmm......
WRT mobile robots, hovering ones are neat, but loose much in practicality
because you use nearly all of your energy just hovering. If you could
stay in ground effect, you could reduce that issue. Still ,unless
you are interested in a certain contest, or the nav issues it is not
as practical.
I still like to have "step counting" capability for nav purposes, as
well. (just as a cross reference.)
BTW, I had considered a 4 motor platform so I could use 2 axis of
optical feedback from a pendulum with some existing motor control
chips. I still think it would work. (replace the pots in a joystick
with shaft encoders. Hang the batteries from the joystick handle. )
Oh well, my ideas for the hovering robot contest. If you use them, and
win, remember to mention me in your speech. (and subsequent royalty
payments :-> )
Good luck!
Alan Barrow km4ba | I've seen things you people wouldn't believe. Attack
jab@hpuerca.hp.com | ships on fire off the shoulder of Orion. I watched
| C-beams glitter in the dark near the Tannhauser gate.
..!gatech!kd4nc! | All those moments will be lost in time -
km4ba!alan | like tears in rain. Time to die. Roy Battyjt34@prism.gatech.EDU (THOMPSON,JOHN C) (06/19/91)
Did anyone see the article about the Moeller flying car in Popular Science last
year? Supposedly Moeller has been working on this flying car which hovers as
well as travels 300mph+. If I recall he had developed a lightweight rotary
engine just for this application. The vehicle was computer controlled and
guided by a single joystick. Supposedly the average person could pilot the
vehicle with little or no training. The article showed an actual working
prototype which hovered with Moeller in it (a crane was used as a safety
net). For mere $100,000 you can reserve yourself a complete vehicle. I believe
he used four engines. For what its worth.
--
THOMPSON,JOHN C
Georgia Institute of Technology, Atlanta Georgia, 30332
uucp: ...!{decvax,hplabs,ncar,purdue,rutgers}!gatech!prism!jt34
Internet: jt34@prism.gatech.edubarts@cyber.Eng.Sun.COM (Bart Smaalders) (06/19/91)
Did anyone see the article about the Moeller flying car in Popular Science last
year? Supposedly Moeller has been working on this flying car which hovers as
well as travels 300mph+. If I recall he had developed a lightweight rotary
engine just for this application. The vehicle was computer controlled and
guided by a single joystick. Supposedly the average person could pilot the
vehicle with little or no training. The article showed an actual working
prototype which hovered with Moeller in it (a crane was used as a safety
net). For mere $100,000 you can reserve yourself a complete vehicle. I believe
he used four engines. For what its worth.
Ah, the old DiscoJet. Moeller used to teach at my alma matter (UC Davis),
before leaving to work on exhaust systems for vehicles and this flying car.
The flying car work is (I understand) funded by the exhaust systems (you may
have heard of Super-Trapp). There are some interesting technical problems
associated with building such a flying vehicle, most of them having to do
with the control problem:
1. The configuration is inherently unstable (like a helicopter). Since
there is no large rotor to slow the natural frequency of response, it
is my guess active computer control is required to keep it from
flipping over once the vehicle is out of ground effect area.
2. What level of computer control redundancy is required? What does all
this fault-tolerant avionics cost? How will the manufacture ever
afford liability insurance (a substantial component of today's
astronomical light aircraft prices)?
3. In hover mode at altitudes above those where ground effect is useful,
it will be difficult to get a high enough dPower/dTime to make for
effective control. Note that a fixed wing aircraft uses it's KE
(ailerons against the slipstream) to get the higly controllable power
necessary to maintain stability in turbulent air. Helicopters have a
large spinning wing, whose pitch is changed quickly. Here, again,
instantaneous manuevering power is provided by stored energy (this time
the inertia of the rotor) rather than modulation of the prime mover.
The two methods likely to work for the Disco Jet are either vectoring
the thrust (but this is hard to make efficient), or variable pitch
fans/propellors and governed engines.
4. What is the fuel consumption?
- Bartdeano@areyes.com (Dean Carpenter) (06/19/91)
Did you ever consider a combination ground effect and nozzle approach ? That is, use an adjustable ground effect skirt, one that can constrict itself down to a nozzle. I have some vague notion of a device much like the old bulb flash reflectors that could unfold from a pie shaped wedge to the full reflector, but obviously it would need to be more sophisticated than that. Exactly how this thing could be constructed, heh heh, I leave up to you all. Would the concept even work ? Say the plenum chamber is 12 inches across in ground effect mode, so thrust is pushing down over a wide area. In nozzle mode all that air whooshing down would be constricted through a 1 inch (say) nozzle, much like a model rocket engine clay nozzle. -- Dean Carpenter uunet!areyes!deano (203) 531-5007 Areyes, Inc. deano@areyes.com "Wherever you go, there you are" sayeth Buckaroo - across the Eighth Dimension
monty@sagpd1 (06/20/91)
In article <25523@well.sf.ca.us> bkoball@well.sf.ca.us (Bruce R Koball) writes: >Re: small RC hovering models...I saw an amazing example of same at a >conference last fall...fellow had purchased it in Japan...consisted of 4 >small PM/DC motors (the type you see in cheap toys) mounted on a >styrofoam ring about 12" in dia.... each motor had a plastic prop about >4' in dia....control electronics and batteries (about 8 oz. worth) in >the center...stability was accomplished via a small *GYROSCOPE* (no >kidding!) in the center pod...gyro had small, dual-gimbeled flywheel >with some sort of proportional sensor feedback (don't recall the >method)...gyro parts (except the flywheel and bearings) were >injection-molded plastic....owner said the whole thing cost about $400 >US....I might be able to track him down if anyone's desperately >interested....reply by email > This discussion of hovercraft has made me think of a possible way to increase stabilty. What if some or each of the "props" that are supplying the air pressure were sort of wheel shaped. The spokes would be the lifting/pressure blades and the outer ring would act like a gyroscope. This would make banking difficult ( unless the whole motor- blade assembly was gimbled), but could not steering be accomplished by vectored thrusts? Also if the vehicle had a "skirt" that was inflatable could you not use helium as the inflating gas to produce additional lift? I don't propose a huge gas bag, just a compromise to add additional "lift". In a rc vehicle you would run the props at a constant speed and use servo controlled air valves to vary the amount of lift. For that matter it might be better in a full size vehicle to run the props at a constant speed to aid in noise and smog emmision control. The props could be variable pitch in a full size vehicle or a venting system could be used. Just thoughts, Monty Saine
tom@intran.UUCP (Tom B.) (06/21/91)
lots of talk about encoders, here is one I used written in postscript.
I wrote a simple postscript
program to create a 100 line disk.
%Change to inches
72 72 scale
%setup a new line moving to center of page
newpath .03 setlinewidth 4.25 5.5 translate
%define number of lines
/lines 100 def
%Angle increment
/angleincr 360 lines div def
lines { 0 2 moveto 0 -2 lineto stroke
angleincr rotate
} repeat
%draw white circle in center
newpath
.01 setlinewidth
0 0 .25 0 360 arc 1 setgray stroke
%eject page
showpage
%all done
I put this paper disk on a piece of 1/16" styrene. The disk was glued
to the wheel using the silicone bathtub sealer. Centering isn't
terribly critical, but I have a circle indicating where to cut the
hole for the axle.mccool@dgp.toronto.edu (Michael McCool) (06/23/91)
I've been following this line for a while, and thought I would put in my two cent's worth of ridiculous suggestions. Would it be possible to combine the gyroscope and the lifting blade? My idea: make the lifter a *ring* that rotates around the *outside* of the robot; possibly make the whole robot T-shaped with the batteries at the bottom of the T for weight balancing. The combination of gyroscopic effect and low center of balance should keep the whole thing stable, and using the blade doubling as a gyroscope will save weight. To avoid counter- rotation, you would actually need two blades rotating in opposite directions. If you could control the speeds of these independently you would have a quick way to rotate the entire robot. The rings would need to have different numbers of blades to avoid "beating" with one another and making a lot of noise. For safety, you would have to either cast the blades out of rubber or encase them in a shield. For forward motion, just have some way of shifting the center of gravity off-center, which will tilt the robot forward (actually, you'd have to deal with the gyro effect, of course... have fun). So, you don't need a thruster. If you had the weight-shifting under computer control you could probably do dynamic balancing and not need an ungainly T shape. You are going to need some pretty sophisticated navigation anyways for a powered hovering vehicle. The robot would end up looking like a flying saucer. I tend to like this because it leaves you a lot of protected space in the middle of the beast to mount your electronics, and a way to suspend things (like cameras and arms) from the center of the robot. By the way, I thought of this when I was 15 and it just came back to me. Don't flame me if it's mechanically impossible; I'm a computer engineer, not a mechanical :-) Michael McCool.
jonker@amc.uva.nl (06/24/91)
In article <1991Jun20.155520.12703@sagpd1>, monty@sagpd1 writes: > This discussion of hovercraft has made me think of a possible > way to increase stabilty. What if some or each of the "props" that > are supplying the air pressure were sort of wheel shaped. The spokes > would be the lifting/pressure blades and the outer ring would act like > a gyroscope. This would make banking difficult. have three of these on each corner of a triangle. Trim the machine to have just enough 'lift' for hoovering within the groundeffect range. > In a rc vehicle you would run the props at a constant speed and > use servo controlled air valves to vary the amount of lift. For that > matter it might be better in a full size vehicle to run the props at > a constant speed to aid in noise and smog emmision control. add a fourth engine in the middle with just the spokes (ie. a normal propellor) that can be tilted. No need for comples nozzles, or what soever. This is the only engine that can be regulated to maneuver in all directions. Seems to me the whole object can be steered by just one joystick, that is, if you can compensate the rotary forces of the center engine. By stationary fan blades mounted in the duct? Catches? ard
jm59@prism.gatech.EDU (MILLS,JOHN M.) (06/24/91)
In article <1991Jun23.095642.872@jarvis.csri.toronto.edu> mccool@dgp.toronto.edu (Michael McCool) writes: >I've been following this line for a while, and thought I would put in my >two cent's worth of ridiculous suggestions. > I have a one of those [suggestions] myself: For a platform hovering in the ground-effect range, you should be able to provide passive stabilization: (1) Break up the plenum under the platform into sections around the edge, or implement it as "pads" at several points around the edge, kind of like feet, and _isolate_ the air supply to the various pads either: (a) use separate air supplies, or (better) (b) bleed the air to each plenum through a separate restrictor from a central fan. As the platform tilts, the exit areas around the low pads are reduced, and the areas around the high-side pads increase. The idea is to make this inherently reduce the pressure at the high corner(s), so they won't lift further, and increase the pressure under the low corner(s) so they won't sink further. EE types can sketch this as a resistor network. The restrictors between the plena _are_ needed. Trimming the pressure in the various pads should allow the platform to float level, even its CG is off-center w.r.t the pads. (Pad _area_ may also be adjusted to balance an off-center load.) This should make the system inherently stable, so little or no control will be needed for leveling. Always easier to try to use the physics of the situation than to fight it, no? Now _my_ ridiculous question: How could one keep the thing from simply skidding off downhill? I don't think this depends primarily on the platform's horizontal attitude, though the "high corner" pads _could_ in principle add a [small] directional thrust component. Anyone have a simple way around this? >For forward motion, just have some way of shifting the center of gravity >off-center, which will tilt the robot forward (actually, you'd have to >deal with the gyro effect, of course... have fun). So, you don't need >a thruster. If you had the weight-shifting under computer control you >could probably do dynamic balancing and not need an ungainly T shape. >You are going to need some pretty sophisticated navigation anyways for >a powered hovering vehicle. The robot would end up looking like a flying >saucer. I don't think this works very well for "hovercraft" type gadgets -- Thrust vectoring sounds better to me, except maybe for a helicopter. I may be wrong, since I haven't built one. That's why I asked. Thanks for any comments. -- MILLS,JOHN M. Georgia Institute of Technology, Atlanta Georgia, 30332 uucp: ...!{decvax,hplabs,ncar,purdue,rutgers}!gatech!prism!jm59 Internet: jm59@prism.gatech.edu