Energy Maneuverability Redux

Military aircraft - Post cold war aircraft, including for example B-2, Gripen, F-18E/F Super Hornet, Rafale, and Typhoon.
  • Author
  • Message
Offline

squirl

Enthusiast

Enthusiast

  • Posts: 58
  • Joined: 27 Feb 2005, 08:13

Unread post03 Dec 2017, 08:02

Folks,

I had discussed this project previously with a watered-down approach, but am now soliciting feedback with a full-fledged version. The paper's target audience is a USAF CAF aviator.

Elevator-ride sales pitch version:

Current TTPs do not quantify how pilots create and solve energy problems. I propose that we do so using an 'energy cost function' model. This will validate existing TTPs and expand our debrief capabilities, capturing lessons that were previously impossible to articulate.

Why here, and why now? I have not been able to get adequate feedback within the USAF. Some have provided encouragement, but are busy with their own jobs. The project is too technical for the CAF, and too tactical for Test.

In the interest of feedback, I need your help to give this a read and offer your thoughts! I would be interested in taking the project in any productive direction.

I have used publicly-available performance data. Even so, many figures are masked by extensive computations.

Thanks in advance!
Offline

sprstdlyscottsmn

Elite 3K

Elite 3K

  • Posts: 3609
  • Joined: 10 Mar 2006, 01:24
  • Location: Phoenix, Az

Unread post03 Dec 2017, 15:35

While I am in no position to address the technical accuracy of the paper, I can trust that you preformed the computational aspect correctly, I am very impressed with what it represents. I have found in my own analyses that sometimes you need a maximum pull and sometimes not. Your temperature plots quantify this. Nice work.
"Spurts"

-Pilot
-Aerospace Engineer
-Army Medic
-FMS Systems Engineer
Offline
User avatar

rheonomic

Forum Veteran

Forum Veteran

  • Posts: 629
  • Joined: 30 Apr 2015, 03:44

Unread post04 Dec 2017, 02:29

Looks interesting; would be nice to see some of the math.

Hard to tell without any references in there, but have you looked at the body of work expanding on fighter agility metrics from around the mid 90s?

Have you talked with anyone at AFRL? Could also try to move forward with AFIT. (Maybe TPS as a class project?)

This kind of formulation would also be interesting to support autonomous air combat.
"You could do that, but it would be wrong."
Offline

garrya

Forum Veteran

Forum Veteran

  • Posts: 576
  • Joined: 25 Dec 2015, 12:43

Unread post04 Dec 2017, 05:25

Very interesting :P
Offline

squirl

Enthusiast

Enthusiast

  • Posts: 58
  • Joined: 27 Feb 2005, 08:13

Unread post06 Dec 2017, 12:01

Thanks for the replies!

I've done a little reading on some of those agility metrics. I got the impression they were intended to compare one fighter against another to account for parameters EM doesn't cover (control response time, maneuver cycle time, etc.). Despite the great body of work, I couldn't find much that actually quantified a pilot's decisions, which motivated me to build the proposed model. Rheonomic: it certainly would provide a quantitative model for autonomous decision-making in unmanned wingmen or network-enabled weapons.

If anyone knows someone who would be interested, whether in military, industry or academic circles, I would appreciate it if you could forward the paper along.

A couple quick words on the math: the colors literally represent the sum (integral) of induced and parasitic drag along each path to the points in question, and account for changes of airspeed (and G available) enroute. There is certainly a path-dependency that can alter those values, and I've noticed that the white interior of the turn circle is never a perfect circle. If this project were to progress to real-world application, there would need to be some ground rules on the methodology. I hold, however, that the overall paradigm is long overdue in quantifying energy problems in air combat.

Coding was initially done in MATLAB, and later in Excel VBA to enable work on any computer. The Red Flag example is actually 100% Excel, with a script that wrote values to a bitmap (later georeferenced in Falconview). I could offer up some of the code if folks were interested.

Interesting footnote: a civil application could be for glider pilots, where the cost function represents the altitude loss required to reach each point at a certain speed.
Offline

charlielima223

Forum Veteran

Forum Veteran

  • Posts: 874
  • Joined: 12 Jan 2014, 19:26

Unread post06 Dec 2017, 22:02

SMSgt Mac did a similar article albeit smaller

http://elementsofpower.blogspot.com/201 ... ility.html
Offline

sprstdlyscottsmn

Elite 3K

Elite 3K

  • Posts: 3609
  • Joined: 10 Mar 2006, 01:24
  • Location: Phoenix, Az

Unread post06 Dec 2017, 23:32

squirl wrote: I could offer up some of the code if folks were interested.


I wouldn't mind having a look at the code for excel.
"Spurts"

-Pilot
-Aerospace Engineer
-Army Medic
-FMS Systems Engineer
Offline

SnakeHandler

Forum Veteran

Forum Veteran

  • Posts: 718
  • Joined: 01 Jul 2007, 18:22

Unread post07 Dec 2017, 00:02

Squirl,

Your next step would be to download the one of the 3-3s, say the Viper edition, and find a way to use your charts to show what's being described within the BFM sections. For example, in the HABFM neutral merge section, use your energy cost comparison to show the why behind the reason an aggressive Out-Of-Plane maneuver works best when an adversary flies out the back of your turn circle. An animated graphic that shows your charts from merge-10s to WEZ would be ideal.
Offline
User avatar

rheonomic

Forum Veteran

Forum Veteran

  • Posts: 629
  • Joined: 30 Apr 2015, 03:44

Unread post07 Dec 2017, 05:51

sprstdlyscottsmn wrote:
squirl wrote: I could offer up some of the code if folks were interested.


I wouldn't mind having a look at the code for excel.


Where as the controls guy, I'd go for the MATLAB.
"You could do that, but it would be wrong."
Offline

squirl

Enthusiast

Enthusiast

  • Posts: 58
  • Joined: 27 Feb 2005, 08:13

Unread post16 Jan 2018, 13:10

Warning: do not download the Excel file (macro-enabled) and change it to .xlsm unless you're comfortable doing so!

For those looking for code, I dusted off my Excel product. It's not very user friendly, but I highlighted all the user-input cells in blue. I haven't commented the code either. The most important thing is to give it a file path to write the .bmp (see spreadsheet for yellow cell). You'll see information duplicated, reminders I left myself when I changed values.

It will give you a "raw" output bitmap of an LFE-style setup, and you have to interpret the colors. For example, the default values in the sheet you'll download have a B-1 running southbound from a Flanker to its north, and escort 30+ miles off to the east. You can see the B-1's position in the picture by the twin 'min-radius' white circles. As you'd expect, turning east is best for the blue team's collective energy state. The Flanker flies pure pursuit on the Bone; the Eagle flies pure pursuit on the Flanker.

Making some assumptions about shooters and missiles, you can see the outcome in the attached image. A 256x256 image took nearly 7 minutes on my computer - it had to run over 65,000 engagements to give you a color for each pixel. Recommend starting small (24x24) and increasing pixel count by multiples of 12. Remember, in this case the colors represent the difference between the bandit's and escort's energy cost once the B-1 arrives at each pixel (it's assumed to arrive there intact). Blue means the B-1 has solved escort's energy problem (dragging the Flanker) while creating one for the bandit. Red means that this didn't happen.

If you're skeptical about opening a macro-enabled Excel file, I might find a way to send you the raw code so you could import it yourself.

Figured I'd release this into the wild rather than delay it further to polish its look and feel.
Offline

sprstdlyscottsmn

Elite 3K

Elite 3K

  • Posts: 3609
  • Joined: 10 Mar 2006, 01:24
  • Location: Phoenix, Az

Unread post16 Jan 2018, 13:34

Wonderful! Thank you!
"Spurts"

-Pilot
-Aerospace Engineer
-Army Medic
-FMS Systems Engineer
Offline

southernphantom

Elite 1K

Elite 1K

  • Posts: 1043
  • Joined: 06 Aug 2011, 17:18
  • Location: Kentucky

Unread post16 Jan 2018, 17:54

I sort of see what you're going for here. I've done similar VBA work, but that was non-simulational and geared towards generating cut and fill cross-sections for a copper-cobalt mine.
I'm a mining engineer. How the hell did I wind up here?
Offline

squirl

Enthusiast

Enthusiast

  • Posts: 58
  • Joined: 27 Feb 2005, 08:13

Unread post13 Feb 2018, 23:25

Folks,

I added some tricks to allow creation of a larger quantity of images at adequate resolution - all to enable animations of this principle over time. See attached pictures and video.

Unfortunately, I have not yet figured out how to superimpose images in a video, so you'll have to use the "initial conditions" image to visualize where the players are throughout.

Scenario: MiG-1 heads south on the western edge of the NTTR, as an F-15 escort heads north on the eastern side. The B-1 is frozen in space to show a range of "topographies" possible when passing that position westbound. Its westbound vector of Mach 0.85 is, however, applied to the computations. Open-source performance figures and notional BVR missiles used.

Remember: the colors show, upon the B-1's arrival at each point, the difference in energy cost between the MiG shooting Bone and the Eagle shooting the MiG. Blue is good and red is bad. For the fighters, the assumption is that both make a max performance turn to get aligned with their targets from each starting position.

Results:

-the B-1's "optimum" escape direction is left-hand initially, but becomes a right-hand turn at around 2 seconds
--this is evidenced by the nearest blue on the B-1's two turn circles (approximated by the chaotic red smear with white dots)
-the Bone's immediate neighborhood becomes more blue as the Eagle passes by, but...
-the Bone's region returns to red as the Eagle flies off to the north
--reason: the Bone's vector (westbound) is still biased more toward the MiG than it is to the Eagle
-a "blow-through" region appears from 2-6 seconds
--it's the model's way of saying that if you can cross the intervening red region, you've adjusted the energy differential by putting the MiG in a tail-chase
--still not a good place to be, and it doesn't help much, since the F-15 is now also in a tail chase, and the region disappears entirely around 6 seconds

Bottom-line: I strongly believe that visualizations like this go further than any other tool the CAF uses to debrief missions. This paradigm does more than any other model I've seen to get aviators to think outside their cockpits and consider the impact of one jet's maneuvers on the strike package as a whole. A lot of programs show playback of all players with airspace boundaries, but against a black background - wasted real-estate in my opinion.

I'm hoping eventually to expand this analysis to an entire LFE vul, and you can see that I'm well on my way. The third dimension is also waiting to be tackled. I'll get off my soapbox for now, and I look forward to your feedback. As always, any assistance or suggested vectors are welcome. Snakehandler: I'm still trying to figure out how to run the computations you suggested. We can talk offline if you have the means to send data with an example of the maneuver.
Offline

squirl

Enthusiast

Enthusiast

  • Posts: 58
  • Joined: 27 Feb 2005, 08:13

Unread post17 Feb 2018, 23:23

Now with rotating CAPs to show the impact of Gomer and escort's headings. Both fighters start on a heading of 090, and fly a complete right-hand circular orbit with a radius of 2NM. In other words, Gomer is initially hot, and escort is cold (see picture).

Initial Conditions.png


I'm sure much could be said about how wingmen & multiple formations would change this simple setup, but I'll let the video do the talking.
Offline

squirl

Enthusiast

Enthusiast

  • Posts: 58
  • Joined: 27 Feb 2005, 08:13

Unread post20 Feb 2018, 04:29

I'll just keep piling on until somebody says something.

Initial Conditions.png

-I moved the striker further east (blue square) and placed big black dots for the red and blue fighters. It's actually easier to see them on the SD video. Seeing this motion makes it easier to visualize how a fighter's hot/cold status alters the hot/cold kinematics of the airspace.

-The arc of smaller black dots delineate the extent of the red fighter's WEZ. It's more than a simple WEZ: it factors in all players' vectors as the striker arrives at each destination. In other words, at a given moment, all destinations east of that line are safe. Notice how this boundary also changes with respect to the red fighter's maneuvering.

Observe that red/blue can happen on either side of the WEZ:
-red airspace on the "safe" side of the WEZ means the striker is unsupported and may soon be threatened if he doesn't maneuver WRT the package
-blue airspace on the "unsafe" side of the WEZ means that this is a candidate vector of attack, and escort may be in the process of opening a corridor here

Either realization should be of use to a force package.

Would appreciate any inputs or questions if this is unclear.
Next

Return to Modern Military Aircraft

Who is online

Users browsing this forum: No registered users and 13 guests