Author Topic: Let your imagination fly...  (Read 1785 times)

0 Members and 1 Guest are viewing this topic.

Offline GrooveHolmesTopic starter

  • Jr. Member
  • **
  • Posts: 43
  • Helpful? 0
Let your imagination fly...
« on: February 17, 2011, 05:43:29 PM »
So myself and a group of fellow RC airplane enthusiasts are trying to probe the feasibility of a rather bizarre but excitingly titillating project.

You'll really need to open up your imagination here as this is really unconventional...

In the realm of large replica RC aircraft (wingspans in excess of 6ft), we're tossing around the idea of what it would take to create a small (<60cm length, <6cm diameter) IR or Optically guided Air-to-Air missile with limited tracking ability. (We'll start with a ground based version)

The idea is to create a replica AA missile built from similar materials as hobby rockets, and using small hobby rocket motors, that can be fired from one RC aircraft at another RC aircraft, track it, adjust the missiles course, then pop the parachute when the missile comes within 20m, give or take, to the aircraft denoting a successful 'kill'.

We have very low expectations of performance, but if we could even in the slightest, give the missile a hint of 'intelligence' to its flight path it would gloriously add to the realism of RC Fighter Aircraft.

At first this sounds really hard, but if you know how modern IR and optically guided missiles work, it seems 'possible' that something similar could be recreated with a small microcontroller, some spatial sensors, an ultra sonic sensor for proximity (to trigger the shoot), micro servos to controll the rocket fins, and some kind of object tracking sensor which would be the hardest part, and why I'm here looking for ideas.

This is not a BAE or Raytheon project and is ultimately being done solely for shits and giggles and pure theatrical sport. Expectations are low, but the budget is sound at about $500 per re-usable missile.

I've pondered the performance of a small CMOS camera to simply track the black object against the blue/white sky and compare frames to make course adjustments, but doubt the FPS performance would be adequate. The missile will boost for only 2-3 seconds before deploying it parachute by default. How far it travels will depend on the weight and drag of the final design, which are obviously yet to be determined. But we're shooting for a range of less than 75m.

The other option is again, CMOS type of camera, but one that can sense IR frequencies, and again track the 'enemy' craft using its engine thermal signature.  These aircraft use Kerosine powered Turbine Jet engines which give off a substantial amount of heat (much like a full sized turbine jet engine).

Details of the internal working of the Fire Control System aside for a moment, the processing chain would work something like this:
1) Pilot aligns his aircraft with enemy craft, putting the viewing area of the missile on to the target aircraft
2) If the missile detects an object in its viewing area, it sends a signal to the microcontroller which sends a signal back to the pilots radio in the form of a audible tone. High tone for a 'lock', low tone for 'no-lock'
3) If the pilot gets a good lock and feels good about the shot, he triggers a switch on his remote which goes back to the plane commanding the FCS to initiate firing the missile
4) once the missile is launched its desired to 'track' the target as much or as long as possible before the rocket motor burns out and detonates the parachute charge (2-3 seconds), breaking the aerodynamics of the missile causing it to fall safely to the ground (much like any other hobby rocket)
5) if the rocket closes within a specified range of the target craft, it has to initial a self destruct mode (popping parachute) to prevent actual contact with the target craft


The only part I'm having a hard time with is the tracking sensor. What sensor could be small enough, and fast enough to meet this need? :)


Into the maths of it...

If a camera could capture 15fps, over the 3 seconds flight duration, it would only have to process 45frames (give or take). This should be enough, in theory, to track an object and adjust its navigation. But i really have no idea as I have no experience yet working with cameras with microcontrollers.


Lastly, it would be an AxonII powering the FCS (for an idea of processing capability), and most likely onboard the missile too (although we have concerns it may be too wide to fit).


The final option would be to find some kind of funky IR sensor that can see an area with resolution, as opposed to a single IR sensor used for distance.


Misc.
-Sunlight will be a factor - for good or for bad
- Lighter is better
- Less power = smaller batteries
- RADAR is not an option. ;)
« Last Edit: February 17, 2011, 05:48:49 PM by GrooveHolmes »
The difference between the student and the master, is that the master knows what he does not know.

Offline rbtying

  • Supreme Robot
  • *****
  • Posts: 452
  • Helpful? 31
Re: Let your imagination fly...
« Reply #1 on: February 17, 2011, 06:31:09 PM »
Have you taken a look at the Wiimote IR Camera?

It can track 4 points of bright IR light within the field of view of the sensor.  If the engine doesn't produce enough of it, you could try adding 940nm LED's on the plane itself.

Offline GrooveHolmesTopic starter

  • Jr. Member
  • **
  • Posts: 43
  • Helpful? 0
Re: Let your imagination fly...
« Reply #2 on: February 19, 2011, 02:25:55 PM »
Hadn't, but will now! Thanks for the tip. At first glance this looks like it would work nicely!
The difference between the student and the master, is that the master knows what he does not know.

Offline klims

  • Robot Overlord
  • ****
  • Posts: 190
  • Helpful? 2
Re: Let your imagination fly...
« Reply #3 on: February 20, 2011, 06:16:15 PM »
Just a though, how would you differentiate between the sun and a plane with IR? I've had problems with this in the past.

Maybe a combination of IR and a low res, low frame rate camera would work. IR would be responsible for high res tracking while cheap CMOS camera could confirm your tracking the right object.

Offline hoosier122

  • Full Member
  • ***
  • Posts: 84
  • Helpful? 0
Re: Let your imagination fly...
« Reply #4 on: February 23, 2011, 11:13:14 PM »
The first missles and torpedos were fly by wire. Just an idea.

Offline GrooveHolmesTopic starter

  • Jr. Member
  • **
  • Posts: 43
  • Helpful? 0
Re: Let your imagination fly...
« Reply #5 on: February 24, 2011, 04:30:48 AM »
And a really good one at that...

That would remove an enormous amount of design work and hardware/weight on the missile, and given the short range of use, a 100m or less spool of cable would probably be manageable with a well designed system.

Thanks for a stellar idea!
The difference between the student and the master, is that the master knows what he does not know.

Offline hoosier122

  • Full Member
  • ***
  • Posts: 84
  • Helpful? 0
Re: Let your imagination fly...
« Reply #6 on: February 24, 2011, 09:34:13 AM »
And a really good one at that...

That would remove an enormous amount of design work and hardware/weight on the missile, and given the short range of use, a 100m or less spool of cable would probably be manageable with a well designed system.

Thanks for a stellar idea!
There wouldn't have to be too much weight on the missile. All of the controls would be on the launching airplane. The missile would just have thrusters and the signal sent through the wire would tell which thruster to burn.

To be honest, the easiest thing would be to fire birdshot or buckshot from your airplane out of a .410 or 28 gauge shotgun. Then you wouldn't have to have a direct hit.

Offline HDL_CinC_Dragon

  • Supreme Robot
  • *****
  • Posts: 1,261
  • Helpful? 5
Re: Let your imagination fly...
« Reply #7 on: February 24, 2011, 04:28:02 PM »
There wouldn't have to be too much weight on the missile. All of the controls would be on the launching airplane. The missile would just have thrusters and the signal sent through the wire would tell which thruster to burn.
Better to use the micro servos on the fins from inside the fuselage of the missile. That whole package could be VERY tiny!(And lightweight)
To be honest, the easiest thing would be to fire birdshot or buckshot from your airplane out of a .410 or 28 gauge shotgun. Then you wouldn't have to have a direct hit.
The whole point in this is to NOT actually kill the target :P Plus I picture absolute horror as a group of spectators gets peppered by buckshot fired from a small aircraft :P


One concern I have with a FBW(Fly by Wire) design is making it reusable AND having the wire safely detach from the source aircraft after firing. I suppose some tiny and loose bullet connectors would work though. That or use tiny spring coils and just shove the control wires into the springs(same concept those old electronics labs from radioshack use) so they get pulled out when the 100m runs out or w/e. That would just be a pain in the butt though to rewind 100m worth of wire heh.
« Last Edit: February 24, 2011, 04:37:18 PM by HDL_CinC_Dragon »
United States Marine Corps
Infantry
Returns to society: 2014JAN11

Offline hoosier122

  • Full Member
  • ***
  • Posts: 84
  • Helpful? 0
Re: Let your imagination fly...
« Reply #8 on: February 25, 2011, 01:18:18 AM »
Ive never priced it, but Id think FBW would be rather cheap and could be one-time use....look at wire for tazors.

You could rig something onto an open-face fishing reel...or re-engineer it out of lighter materials.

Missles or buckshot...spectators should not be anywhere close...probably in a bunker or hardened shelter.

Be sure to look at the engines used to shoot toy model rockets...maybe even. Fireworks such as bottle rockets or roman candles...regardless it's going to be difficult to track at significant speeds.

Good luck!

Offline 4by4

  • Full Member
  • ***
  • Posts: 61
  • Helpful? 1
Re: Let your imagination fly...
« Reply #9 on: February 27, 2011, 09:02:25 AM »
How far have you thought this through? It's true that air to air missiles such as Sidewinder use thermal IR, but CMOS cameras can't detect thermal IR. You will need to think through and test your sensor technology before you get to the challenges of fast airframe control and maneuver.

Offline Mansoor

  • Jr. Member
  • **
  • Posts: 20
  • Helpful? 0
Re: Let your imagination fly...
« Reply #10 on: February 27, 2011, 12:36:35 PM »
In a range 2 to 3 sec. whatever control logic you have on that rocket must be very very fast. how will u do image processing that fast? super fast micro controller? one way i can thing of for high speed data processing is using an FPGA. but what do i know. Also doesn't the parachute open at apogee normally. At that point there isn't much pressure on the cone to keep closed. what happens if the rocket comes close enough to open the chute but the speed is too high for the cone to open, so the chute charge explodes but the cone stays on -- kaboom X_X dead plane. assuming good aim  ;)

whatever you do though please shoot videos of it and post online. this should be awesome.

Offline 4by4

  • Full Member
  • ***
  • Posts: 61
  • Helpful? 1
Re: Let your imagination fly...
« Reply #11 on: February 27, 2011, 02:45:47 PM »
Hmm, I've been thinking about this all day, because it reminds me of years ago when I did some experiments for an idea - to have a model rocket return to a specific point on the ground to simplify recovery. They are similar in problems of detecting, tracking and homing. Based on that, I think homing to a beacon simplifies a lot of things. You could think of it as simulating the thermal signature of the jet engine in the optical spectrum, which allows you to use an optical sensor. If the beacon is a certain color, you can use existing color tracking algorithms in RoboRealm.

I think a thermal sensor is feasible, and I have built long range thermal detectors (> 100 ft), but it is a lot trickier. You have issues like the window material for thermal IR, which, if the sensor is in front, needs to be aerodynamic. You also have to make do with one or a few detectors, because an array detector is too expensive.

Locating the complexity off-board of the missile is a good idea, but I think radio control is simpler than wire-guided.

 


Get Your Ad Here

data_list