Military AI Works

Below is a link to an interesting post I read in the FSDeveloper Forums. Does anyone have any other information on this AI V22 Osprey?

http://www.fsdeveloper.com/forum/showth ... p?t=425387

Now that would be a well received addition to the community !


Fingers crossed

And toes now too, the pilot models look familiar.

I wonder if that's just a modification of the AH-64/CH-47 takeoff animation, but instead of rising two or three feet off the ground it goes up twenty... the converts to engine forward and goes racing down the runway.

I was wondering that myself. I was even thinking that maybe they have animated the engine rotation according to forward speed, but the tricky part would be the landing. To do it smoothly you would need to have very good flap control.

It will be interesting to see how it works out.

Not my V-22, but have definitely experimented with the whole issue.

Tieing the rotation to forward speed isn't a problem, ask 'magic' Mike.
It's also one of the variables I will use for Marks F-35B rotating nozzle I am helping him with.

Landing isn't the problem, it's shutdown.

I am fairly sure that when the aircraft goes inactive all variables go to 0.
So although the code works great for active / taxi / takeoff, it's quite ugly for the reverse.

Eg.
Model hover taxis back to its spot.
As soon as it goes dark all the flags / covers re-appear, then it settles back down to the ground.

I have thought of a way around that, but the coding definitely works better if it taxis out and in on the ground.

Same also applies to ALL skidded Helis which HAVE to hover taxi in/out.

Been trying to figure out a way of getting the rotors to stop in the same location on shutdown also, no joy with that yet either.

Is there ANY way of finding out which variables (if any) continue to send data after the aircraft goes inactive, eg is prop rpm still active until it stops?
Or engine rpm, or oil pressure etc etc.

Kevin,

After all the work we did back in the spring of 2007 I thought we were able to get your "proof of concept" KRAI MV-22 Osprey to have decent STOL takeoffs and landings. I just dusted it off and had it fly around the KBTV VFR pattern to refresh my memory as to exactly what we had accomplished back then and what AI FDE improvements I had added to it since and it behaved pretty much as I remembered and expected.

Here is what I observed:

1. When parked its engines were vertical, the port engine had various red covers and warning flags located on it. Fifteen minutes prior to engine start when the nav lights came on the red covers and flags disappeared.

2. After engine start when it started to taxi the engines tilted forward ~5 degrees from the vertical and remained there as long as it was taxiing. Whenever it stopped on the parking ramp, taxiway or runway the engines returned to their vertical position, but tilted back forward ~5 degrees when it started to taxi again.

3. On its takeoff roll when your Osprey passed ~30 KIAS the engines' tilted further forward to ~15 degrees and it lifted off at ~69 KIAS with a short takeoff roll of ~390 feet.

4. During its climb out the engines tilted further forward to ~45 degrees as it passed 75 KIAS and to ~90 degrees or horizontal at ~170 KIAS.

5. As it entered the VFR pattern and slowed down past ~170 KIAS the engines tilted back up to ~45 degrees and remained there on the ~80 KIAS downwind leg.

6. Your Osprey remained at ~80 KIAS with an ~45 degree engine position on its base leg and as it turned to final it slowed below ~75 KIAS and the engines tilted further up to ~15 degrees.

7. It flew a nice stable final leg at ~44 KIAS and an engine position of ~15 degrees forward of the vertical.

8. It slowed to ~31 KIAS as it touched down and then immediately stopped, before taxiing in at ~20 KIAS with the engines ~5 degrees forward of their vertical position.

9. It parked nicely and the engines shutdown in their vertical position.

10. Fifteen minutes after engine shutdown the nav lights went out and the red covers and warning flags reappeared on the port engine.

In short, it performed pretty well as a STOL AI model.

Personally I don't think hover taxiing should be an issue with airport based KRAI MV-22 Ospreys. The videos I've seen all show them taxiing out normally and onto the active runway before using either their STOL or VTOL capability to depart and then after either a STOL or VTOL landing on the active runway, taxiing back in normally.

Agree with Mike, just had 2 down here in TX training and got to see them fly by but in helo mode, surprisingly they never converted to 45 or airplane.

Taxiing out and back on the ground isn't a problem for wheel based VTOL stuff.
You can tie the hover anim to the strobe light, with a delay so its lined up when it does go into the hover

I forgot that AI a/c leave their nav lights on after shutdown, so that solves one issue in that a taxi hover out and back from the runway can be tied to the beacon light.
However that also depends on how quickly the rotor spins down.
The trigger has to be low enough to get it into the hover before it starts taxiing, but high enough to stop weird stuff happening during final.
I.e. when the 'rotors' slow down because they aren't constant speed.

Another problem is if you look at the wheels when they are in the hover and moving forward, they still turn.
There is a sort of fix for that which is to hide the wheels when it goes into a hover and replace them with dummy ones that only show at that time. Still not perfect as you can see them change if they dont 'line up' with each other, as they will appear to move when the dummy wheels appear/dissapear.

The bigger issue is skidded helis!
Anything that works similar to the above solution makes it only suitable for drive thru parking, you can get away with a wheel based one getting pushed back.

The model shown over at FSDeveloper is from James, the author of the posting. I have pointed him to this thread and asked him to say hello and share his experiences if he feels like it.

Weescotty wrote:The trigger has to be low enough to get it into the hover before it starts taxiing, but high enough to stop weird stuff happening during final. I.e. when the 'rotors' slow down because they aren't constant speed.

I am not quite sure what you mean here. Are you talking about the speed that the blades turn at, i.e. visually? If so there is a simple fix for that. I worked it out when dealing with older prop models that have tick over anims at low power settings. It does mean juggling a couple of other parms but Mike can do that.

That's exactly what I am talking about.
I thought the rpm levels for the 3 states were fixed in FS4?
I know visually you can replace the middle one with a copy of the fast blurred disc, but that wouldn't fix the underlying issue.

OK then we may be able to overcome that. I did with Dave Friswell's PC-9, I think.
What I found is that if you set the min_throttle_limit, under GeneralEngineData, to at least 0.08 so 8% throttle then the props/rotors stay a blur rather than the unsettling look of an engine failure. Double check the limit but I don't remember having to go as high as 0.1 .

It would not affect taxying as aircraft use 10% anyway, but you have to make sure that the brakes can stop quickly enough with power still being applied and obvious hold easily with up to 10% power applied.
The other main thing it will affect is the aircraft deceleration whilst in the air, although it may actually counteract an aggressive deceleration fde. If not you need to dirty the plane up to offset this but all this will be child's play for Mike.

I would say give it a try to see if it helps that particular problem.

Keep working on it, boys! I sure would love to have some Ospreys in my sim!

Goys,

It appears that James is going to release his AI V-22 Osprey as Payware.

http://www.fsdeveloper.com/forum/showth ... p?t=425363

Mike if you would like to post a few pics of mine go ahead.
It maybe getting about time to have a fresh look at it.
I have learnt a lot since 2007, and I am sure I can add some tweaks to it. There were a couple of part I wasn't 100% happy with, but I now have the skills to improve them.
I also lost the FDE for it!

Weescotty wrote:Mike if you would like to post a few pics of mine go ahead.
It maybe getting about time to have a fresh look at it.
I have learnt a lot since 2007, and I am sure I can add some tweaks to it. There were a couple of part I wasn't 100% happy with, but I now have the skills to improve them.
I also lost the FDE for it!

Hey Kev, as this thread is moving towards your Osprey version, can you start the discussion up in the Skunk Works please?

Weescotty wrote:Mike if you would like to post a few pics of mine go ahead.
It maybe getting about time to have a fresh look at it.
I have learnt a lot since 2007, and I am sure I can add some tweaks to it. There were a couple of part I wasn't 100% happy with, but I now have the skills to improve them.
I also lost the FDE for it!

Kev,

I'll take some screenshots of your KRAI MV-22 Osprey in action and post them as soon as I get a chance.

I've learned a lot about AI FDE's since 2007 and have applied some of that to the KRAI MV-22 Osprey FDE already, but I can probably do even more if you resurrect the model.

I can also PM you a copy of my current KRAI MV-22 Osprey FDE if you need it.

Guys,

Below are some current screenshots at KBTV of Kevin's 2007 "proof of concept" KRAI MV-22 Osprey for FS9.

It is from early in Kev's AI model development process and as such it does not have all the usual "bells and whistles" built into it yet, but it did and still does show great promise for becoming an outstanding and very unique AI aircraft model.

Enjoy!

Parked on the VT ARNG Ramp


Taxiing Out on the VT ARNG Ramp


Taxiing to Runway 15 with Another KRAI MV-22 Osprey Demonstrating its STOL Capability in the BackGround


Cleared for Takeoff on Runway 15


Lifting Off Runway 15 in Less than 400'


Climbing Out with Engines Tilted Forward 45 Degrees


Leveled Off with Engines Tilted Fully Forward


On the Downwind Leg of a Right Hand VFR Pattern


Preparing for the Turn to the Base Leg


Descending on Final to Land on Runway 15


About to Touch Down on Runway 15


Landed on Runway 15


Taxiing Down Runway 15 to Taxiway H


Taxiing In After Turning onto Taxiway H

There's a big smile on my face, that really does look very good.