MAV Study Focus Group Post

[a_majoor]

I don't know that you need multiple types of MAVs so much as you need mini mission pallets for want of a better word. (These would be factory installed, the CQ just pulls the right one off the shelf and sends it to the DP).

The basic use is local surveillance, so the "standard" package would have a camera with a wide angle lens, and a "flip mode" to a fixed zoom factor (3x ?) for pinpointing the target. Version two would have a thermal imager camera. The only other feature the standard MAV should carry is a laser pointer/designator. Other MAV payloads are possible, but for the Infantry section, Engineer section and AFV commander, these would be the most important.

A fixed wing design that fits into a launch tube, perhaps cylindrical in design with a payload mounted in the nose, pop out wings and tail after it leaves the tube, and a shrouded propeller in the rear would be a flexible airframe that could do "racetrack" orbits over a patrol, fly out to 2000m to spot for the support weapons and fly slowly enough for target observation (although it won't hover). It could also be inexpensive to mass produce.

 

[MarkR]

Hi all. 

We lose one, we gain one.  Allan - thanks for your input.  I'll forward remuneration details when I get them.

KevinB is joining us.  This is a late notice change, but in consultation with my supervisor, we don't think it will cause any undue affects to the program - especially considering this whole focus group is virtual. 

KevinB, welcome.

Just to recap:

WRT to missions/usage, so far we've mentioned:
1.  Recon
2.  Marking or lazing a target
3.  Patrol
4.  NBC detection (through DARPA)
5.  Munitions delivery (albeit limited)
6.  Spotting for platoon support weapons.

WRT to features:
1.  Cameras (wide angle, infra red, zoom)
2.  Some form of switchable payload for different missions
3.  "Hands-off" or simplified piloting scheme
4.  Some form of map on the controller that can display troop positions via standard icons
5.  Map interactivity (likely at HQ level)
6.  Ability to share data to other groups.
7.  Stealthiness
8.  High bandwidth
9.  GPS positioning/autopilot (ie: click on map, vehicle navigates to that spot)

48Highlander brought up a good point as well that different MAVs are better suited to different missions - such as a hovering MAV vs a traditional airplane flying MAV.  You could also include in this list ground vehicles - say a small controllable ball or car that you could throw through a window or drive around a corner to see what's what.  This of course then generates more Human Factor problems, like attempting to keep the interface consistent in both control and operation while at the same time ensuring the soldier doesn't get confused on what "mode" they are in (ie: flying, driving, hovering, etc).

I've likely missed a few things.  These were just off the top of my head.

Keep the ideas coming folks.

Cheers
Mark

 

[KevinB]

Thanks for letting me get in Mark - even if I was tardy.


Control/Display - has any thought been given to interface with the LandWarrior/Future Warrior data systems? The Xybernaut etc.  As well to intergration with the IFF/location setup with the SAS (Situational Awareness System) and display on the C-SAM of a LAV.
 a.) I don't like the idea of voice activated command, if you anything like me you get a little squeaky when scared and it may not react to your words of command - secondly you migth have it do something you did not want, while you .
 b) using the Xybernaut touch screen as an examply you could have it setup for simple commands - forward, back, left right and a second touch for distances.
  c) using the same icons and interopting with the SAS is key IMHO - being seemless intergation to higher - and back.

Flight - I think a hover / loiter w/ tight race track is a necessity - a_majoor makes excellent points as to range etc. however all things being equal the more range the better.
 
Data Transmission - I'll admit I am clueless on the bandwidth issue - would it be easier for it to send stills rather than motion video?  I'm not sure at the section level we need 100% realtime video all of the time - it would be nice - but more importantly I think we need the immediate ability to send info higher - you could turn on video once something is observed

Payload - so much for being able to drop munitions - but perhaps some sort of small magnesium flare  - it could be used to draw attention / fire away from the section just breifly - like a tiny DD? and double as a self destruction means.

Laser Designator / Illuminator - talk to user JD here - he is the SME.  I'd like to see a variable system for pulse and steady beam so if the vehicles are operating in conjuction with others they can easily discriminate.

Vision/Detection Systems - II/FLIR - MAD (can it be made that small?) just an idea on using Magnetic detection for clearing woodlines etc.

 

[48Highlander]

I was toying with an idea, but I suspect it would be too complicated for a simple machine (KISS is a good rule to follow...): fixed AND rotary wing flight, not unlike the US Osprey. Props forward for fast movement forward, props up for hover, slow movement, rearward movement, lateral.......

Don't know how feasible, but it's a thought.

Very feasable, and was done by the US.  The helicopter propeller can be stopped in mid flight and locked in place to serve as a wing, while a jet engine in the center of the aircraft propels it to high speeds.  However, doing this for a MAV would be too pricey, make them too bulky, and isn't really neccesary.  Better solution I suppose might be building a MAV with a large enough wingspan to allow it to fly at very low speeds (like a glider?), yet also be capable of ataining high speeds without tearing itself apart.  I'm not sure about the physics behind that, we'd need an aircraft tech for input, but I'm pretty sure it's doable without too much difficulty.  The other advantages of a longer wingspan would, ofcourse, be longer flight time for the same ammount of battery power.  And you'd have enough surface area on the wings to hook up photovoltaic cells, which would give it a heck of a lot of flight time between recharges.

So ok, assuming the proposed MAV can travel slow enough for detailed target observation, anyone still see a need for a helicopter type MAV?  I'm starting to lean more towards the single airframe, multiple package idea.

 

[MJP]

I think a helo MAV would be benificial in some circumstances, but for the prpose of providing some eyes on for a Sect/Trp/Sqn, a traditional winged airframe is fine.  The MAV itself itself isn't going all that fast so loitering over one spot is relativly easy to do by just race tracking over the position as 48th suggested.

Kev brings up an excellent point in regards to SAS.  I've used the system and was pretty underwhelmed to say the least, however my prejudices apart I think that the MAV could have the abilty to project what it sees onto the SAS using the standard map icons.  I don't see the MAV it self sending this info to the SAS system but the interface itself.

 

[MarkR]

Dumb scientist question here - can you tell me what SAS stands for? ???  I'm pretty sure it's not the Special Air Service.

Thanks
Mark

 

[MJP]

Situational awareness system...overseas we basically use it(right now) to track our vehicles, but it's capable of a whole lot more.

Here is a brief overview here http://www.army.forces.gc.ca/lfwa_hq/feature_situational_awareness_training.htm

 

[KevinB]

Situational Awareness System -- it works in conjunction with PLGR and TCCCS to IRIS and gives units the ability to see where they are - where others are and where the enemy is on a digital map

 

[48Highlander]

SAS - Situational Awareness System

http://www.army.forces.gc.ca/lfwa_hq/feature_situational_awareness_training.htm



One other thing I meant to cover as far as bandwidth issues go - anyone have any idea just how much bandwidth is used up for video transmission?  And what sort of data compression/encryption are we talking about here?  I don't see there being a problem with bandwidth if you utilize a complex compression algorythm.  If you want to go off-the-shelf, there are proccessors on the market capable of DivX compression/expansion.  Assuming a bitrate of 700kbps, you could have DVD quality imagery from the MAV.  Considering that the 801.1G wireless transmission protocol is capable of speeds up to 125Mbps (about 178 times faster if I did my math right), bandwidth shouldn't really be a problem.  You could even have the MAV transmitting ultra high quality pictures of the target area for recce purposes without creating a bandwidth problem.  Ofcourse, maintaining that speed over a distance of 2km might prove difficult, however, no matter how slow the transmission speed, I can't imagine it dropping below 1mbps, which would still be fast enough for our purposes.  The bigger problem would be providing enough electricity to power the data transmission over ranges of up to 2 km - which once again would make solar panels pretty cruicial unless you only need your MAV for a few minutes at a time.

 

[a_majoor]

Moving back to the physical as opposed to the electronic, a tube launched MAV could have a "cruise" and "loiter" mode; the wings snap out to 450 on launch for cruise, and extend fully to 900 for long term loiter.

I think 48th has hit on an important point; power supply. Even with Lithium Ion batteries, it will be difficult for a MAV to have long endurance and send high bandwidth transmissions over extended ranges without running into serious power supply issues. Perhaps the MAV needs a traditional model airplane engine to run things (2 or 4 cycle engines with glow plug ignition using methanol fuel or diesel). There are some medium term work arounds, though. The MAV could actually be powered by a "micro turbine" engine; some of these prototypes are the size of a dime, and are designed to power laptops and other electronic devices for extended periods of time. Fuel cells that can work with a liquid fuel are also in the prototype stage (handling hydrogen in the field is a headache I don't want the troops to be saddled with). The only reason I don't mention solar cells is the MAV needs to be able to work 24/7, including when the battlefield is obscured by dust, smoke or fog.

 

[Grunt_031]

power supply

Maybe a Hybrid type engine both conventional/battery/solar panel?

as for range. how would it fit into the ISTAR plan/ Surveillance plan. I think the 2 km or less at the section level would be good.

Are they not have problems in Iraq with the problem of coordination of all the UAV/Aircraft.

 

[Grunt_031]

Here is a good link discussing the same ideas we are hashing.

http://www.defense-update.com/features/du-2-04/mav-design.htm

 

[MarkR]

Yeah, that's a good link.  I've visited that quite a lot over the last few weeks or so.

Okay all, you've given me a lot of good information to go by.  There's just one last thing I want to ask you, and I know it's been partly covered before, and that's more detail on the actual controller interface itself.

Specifically I'm looking at how you want to control this thing.  Previously, it was mentioned that most of the actual flying should be done automagically, leaving the soldier to either watch the video feed or keep eyes on the surrounding terrain.  Therefore, assume the MAV carries enough onboard intelligence (or in the controller) to fly itself to wherever you tell it to, and can avoid obstacles.  Also assume that all the technical problems that have been discussed have been solved.  You've got in your hands a 15 cm flying machine that can hover or fly a racetrack pattern, with a range of up to 2 km, and an endurance of up to 1 hour.  Current payload is wide angle and zoom cameras.  You've got a large PDA device (or laptop if you prefer) as the controller.  You've got a village or wood to recce and there may be unfriendlies in the area (always design for the worst case scenario). 

You know roughly where you want to put the machine.  What do you think is the best way to tell it where to go? 

IE:
1.  Have a map display and click in waypoints with airspeed, altitudes, heading, etc.?  Then sit back and watch the camera feed.
2.  Have a bunch of "knobs and dials" to set heading and altitude for a semi-manual piloting capability?
3.  Have the ability to switch back and forth between the two as mission requires?
4.  Some other kind of input scheme?  Voice? Joystick?

This is my last question, so please everyone sound off on it.  After this, I'll be writing up a Product Design Specification based on what you've talked about and on my research into MAVs which I'll email to you if you wish.

Mark

 

[KevinB]

#1 - with touch screen buttons for additional input (like right/left, forward/back, up/down and loiter)
Something like the xybernaut

the thing on the wrist...

 

[Grunt_031]

1.  Have a map display and click in waypoints with airspeed, altitudes, heading, etc.?  Then sit back and watch the camera feed.
2.  Have a bunch of "knobs and dials" to set heading and altitude for a semi-manual piloting capability?
3.  Have the ability to switch back and forth between the two as mission requires?
4.  Some other kind of input scheme?  Voice? Joystick?

1.    I think that is the easiest may to go, the preflight planing would have to be 3-D (Hopefully via Sat imaging) to take advantage of the tactical movement of the MAV. If it did have the AI to follow the map of the grd without instuction would be great and eliminate the need of the controller to program it.
2.    semi-manual piloting should be done via touch screen commands, image feed adjusted by a small joystick.
4.    Voice command maybe a option in future but the ones we tested were unreliable and hard to calibrate. I think that this would be the least desirable of all the input option because too many things could go wrong with it. Joystick for the flying or movement would take alot of practice to use, restricting a joystick to a secondary function such as camera manipulation would be alright.

The size of the interface controller should be about the size of the xybernaut, or smaller. The restriction is , a smaller device is that it must have great resolution for the imagery feed. These days the smaller the screen the less you see, or the quality is poor.

 

[a_majoor]

I would not want to be glued to a controller (especially if I had to move suddenly), so perhaps a map "pre planning" session (set a series of waypoints for the MAV to follow), and a means to override if you identify a target. This needs to be quick and intuitive. I am going to assume a setup similar to KevinB's picture on the previous post:

Step 1. The display shows a map, and the operator inputs the waypoints by tapping the screen with his stylus (or fingertip). (If a 2-D map is used, the MAV may have a default altitude, or be able to assume a constant altitude relative to the ground.) If the MAV is recoverable, it might be simple for the control unit to have a "come home" homing signal or beacon, allowing the operator to recover the MAV anytime on the move. One toggle switch to initialize the MAV and recall it

Step 2. The MAV is initialized and launched. The display goes over to the camera feed (another toggle switch to allow back and forth between the image and map) (third toggle switches between wide angle and Zoom)

Step 3. The controller sees something on the screen which arouses his interest. He taps the stylus on the image twice, and the MAV assumes an orbit which keeps the image centred on the screen. (This would require a fair amount of intelligence built into the MAV, I think).

If I have this figured right, the display and control unit has a low glint screen, a stylus attached by a wire or cord, and three toggle switches on one side:

    Map/Camera
    Normal/Zoom
    Initialize/Return

I still have a bit of an issue with the controller strapped to the wrist; it is an indicator of who is important in the section; the screen's glow could have tactical issues at night, and the resolution of the image might not be there.  Maybe in the longer term this could be done in "virtual reality", i.e. the camera sends images to a monocle display while the actual MAV control is done through a "data glove" which interprets hand movements into controls (mostly for overriding pre programmed manoeuvres). Of course enemy snipers are going to be looking for guys who are waving their hands around spasmodically 

 

[MJP]

IE:
1.  Have a map display and click in waypoints with airspeed, altitudes, heading, etc.?  Then sit back and watch the camera feed.
2.  Have a bunch of "knobs and dials" to set heading and altitude for a semi-manual piloting capability?
3.  Have the ability to switch back and forth between the two as mission requires?
4.  Some other kind of input scheme?  Voice? Joystick?

I think #3 would be the best with an option to change altitude and speed through the interface, of course I'm assuming that to change airspeed and altitude it would not be a knob that you have to turn but rather an interface button.  The problem with a map display/Touch screen/pad that you click in on is that they usually require a special type of stylus and things like that tend to get lost rather quickly in the field.  Plus looking at how small some screens are(not that you want a large one), some people may have "large fingerinitus" and  not be able to use it properly.  I've personally found them to be finicky at times and  especially hard to use with gloves on.  But if technology allows you to use anything to I would prefer a touch screen of sorts rather than manually adjusting or inputting waypoints.

I like the Americans approach which is to make the interface similar to controllers that their soldiers are already use to working with(PS2, Gamecube, gameboy).  Game boy is the picture I get in my mind when you talk about the interface.  Either attached to the wrist or hand carried/'binered to a persons tac-vest or the option to do both.

Simplicity would be the key for controlling the MAV through the interface.  Easy to use Pre-launch, Launched, and Camera menus that can be called up fairly quickly in a fashion similar to a_majoor's suggestion of a double click on the screen idea. 

 

[KevinB]

I just want to clarify that I DONT like the xybernaut method of wearing for this system - more of a size /interface setup idea.  I (personally) dont see this as something for the section commander to play with as he has enough on his plate, and with all the highspeed gizmo's we deploy with these days this sort of thing won't be much of a target identifier.

I'd rather have it in a pouch on your LBE able to foldout (the pouch) and be used both with the pouch 'unstowed' for movement (you moving) and seperate from the pouch - or pouch removed from vest - when static.

As much as I hate to go off on a tangent - this sort of thing for the LI-SOC building block 4 man det could be ideal - as well as for a 4 man (heavy) sniper det to pick a route to the FFP.  it would be nice if the MAV could accept (on the fly) digital aerial photo's from other assets.


Cheers

 

[Andyboy]

Hi everyone, I meant to contribute to this discussion before but I just started a new job so I haven't had much time to reply.

When I was in my fourth year my minor project was working with the aerospace engineers at Carleton who were designing a UAV to patrol the coast around NFLD for oil spills. The ID portion of the project was the portable control station which was meant to be housed in a pickup and used at the landing site in case the A/C needed to be landed visually.

We researched a lot of new technology that is available (that was about five years ago) and the technology exists to do pretty much whatever you want to. Soft, foldable/flexible screens, soft keyboards, wireless controlers, the ability to broadcast sound to your ear wirelessly and without anything in your ear, the ability to control simple functions on a computer using thought, screens the size of a thumbnail with resolution equal to any desktop computer...believe it or not, it's all possible. Which leads me to the only point I can really make, why design a requirement based on what you think is possible rather than what you think is required?

Andrew

 

[MarkR]

Hi AndyBoy. 

Designing based on soldier requirements is actually what I'm attempting to do here and is the end product of Phase 1 of my MSc (Development of the Product Design Specification).  I'm a cognitive/human factors specialist, so the actual physical design of the MAV itself doesn't much matter to me, as long as it fulfils the needs of end users and allows me to build an interface without imposing a lot of complicated constraints.  And I agree, given enough time and resources, just about every technology is possible these days.  (I've seen MAVs that look like big mosquitos - complete with flapping wings to fly and the ability to "walk" - the creator of which is currently attempting to make them "forage" for combustible fuel in the natural environment.  How Science Fiction is that?)

Anyway, based on the suggestions everyone has made, the next step is to generate a list of user requirements, then bend the technology to fulfil that list.  Then of course take the whole thing back to the users over and over to "tweak" it, because no-one ever designs right the first time around.

There are still a couple of people I'd like to hear their opinions of concerning the nature of the interface, such as yourself, and KevinB.  Even if you agree with what was already said, please simply say so. 

Thanks everyone
Mark