AI Aircraft SID and STAR Controller

[FlyTweety 2]

Is the first part at a 2x sim rate?  I ask because I notice some snaking after turns at a 2x sim rate.  Also, the beginning looks like the AI is in an area where it should be intercepting the FSX-IAF.  Sometimes the AI has to turn abruptly to intercept it because FSX inserts it out of nowhere (you can see it happening in the FSX Traffic Toolbox Map).

 

That being said, I don't think the AI needs to be in "hot" mode during the whole final.  It is easy for me to add a parameter that turns on fine adjustments below a user-specified AGL atltitude, so I'll add that.  I'm also going to add AITurnScalar specifically for final (also controlled by the user-specified AGL altitude).

 

I wouldn´t die for it but i am sure, that the first part is without any acceleration, means 1x.

Will confirm it tomorrow after trying the same flight again.

You are right, this is right in the IAF area.

But apart from LOWI i experienced that behavior (Rattlesnake"walk") at many places as soon as the AI is assigned to the "Final" File.

 

Would be great to have this additional parameter, this would wipe out some of the "unrealistic" flight patterns where we don´t really need them.

Take LOWI again into account, here we need first the AITurnScalar (i used 1.4 for the "normal" STAR)  to be able to turn 180° in the tight Valley (soft mode), even later the "hot" mode to enter the short final (no further turns needed).

Otherwise, with the default setting, the Valley is to tight!

[ysfsim 22]

(remember, new AI traffic will still get injected parked at a gate - so gate occupancy won't go down that much)

It does?

My test failed. Even though I force the ai to ground in slew, it still says "enroute". Can you find a way to tell fsx "rollout" or something. Hopefully that way it can get taxi clearance.

[Clipper Ocean Spray 189]

My test failed. Even though I force the ai to ground in slew, it still says "enroute". Can you find a way to tell fsx "rollout" or something. Hopefully that way it can get taxi clearance.

 

I do use UT2 mostly, which will re-inject an AI at gate parking that was just deleted after rollout.  Generally, UT2 is very aggressive about re-injecting AI that should be in the reality bubble according to its time schedules, but are not.

 

Regarding rollout, I agree with you, it probably does come down to one variable, but it would need to be hacked (not settable via simconnect).  I'm looking into it, but it is really slow going.  There hasn't been a lot of work by the community in this area (documenting unknown variables within the sim related to AI traffic states). 

[ysfsim 22]

I do use UT2 mostly, which will re-inject an AI at gate parking that was just deleted after rollout.  Generally, UT2 is very aggressive about re-injecting AI that should be in the reality bubble according to its time schedules, but are not.

 

Regarding rollout, I agree with you, it probably does come down to one variable, but it would need to be hacked (not settable via simconnect).  I'm looking into it, but it is really slow going.  There hasn't been a lot of work by the community in this area (documenting unknown variables within the sim related to AI traffic states). 

If that can be broken, imagine the possibilities... You will also have the ability to lead ai around by the nose. Things like runway usage would be up to us. I ultimately want the ai to fly SID and STARS ( done), use runways defined by user created cfg files that reflect actual usage and noise abaitment limits etc.

[mike2060 4]

The program seems to be a lot better since the last time I used it so good job.

 

I noticed that sometimes AI on final will drop suddenly to get on glide path. I assume this is because they finish the STAR at an altitude that is too high for the final approach and they never get a signal to descend to the proper altitude. Is there anyway to use the approach section of the PMDG navdata to find the approach altitude so that at the end of the STAR they are told to descend? So say the STAR's last altitude restriction is 3100 feet but the ILS approach has an intercept at 2600. When the AISIDSTAR hands the AI over the FSX, it tells it to descend to 2600. Not sure if this is viable at all, just throwing it out there. Maybe they do descend already but I seemed to notice that sometimes they were high.

[mike2060 4]

Well I probably spoke too soon. At CYHZ the AI were landing RWY 23 yet some aircraft were getting assigned STARs for RWY05. Any idea why?

[Clipper Ocean Spray 189]

Thanks,

 

 

 

I noticed that sometimes AI on final will drop suddenly to get on glide path.

 

I think this is fixed (or better) in the version I'm about to put online.  The problem was I was using too hot of a vertical speed scalar (gain) during final approach, which was causing stability problems (oscillation).  I've gradually realized I don't need that much gain, so I took it out.

 

 

Well I probably spoke too soon. At CYHZ the AI were landing RWY 23 yet some aircraft were getting assigned STARs for RWY05. Any idea why?

 

 

This is a big fix for the new version.  The program reads weather data at the monitored ICAO to determine landing runway (initially before the program starts relying on actual FSX runway assignments for incoming traffic).  Also, I've noticed that the some of the up-to-date STAR data doesn't match the actual runway data in FSX, which could cause problems (e.g., AISIDSTAR picking a default landing runway 5 because it couldn't match runway 23 in the STAR to its old number 22 in FSX).  This has been fixed as well. 

 

-Roland

[mike2060 4]

Great, thanks a lot. I flew a bit and at least at CYYZ with a ton of traffic and CYUL with a lot of traffic the planes were flying the correct STARs.

[pberg 0]

Thank you for such an awesome program!! I succesfully created a SID for TNCM (St. Maarten) so now aircraft are vectored south after takeoff like they should to avoid the 1000ft hills. Thanks again!!

 

Patrick

[bib974 0]

Hi Roland and everyone,

 

Now.that the.forum is.back online, is there any news about the latest version? Ir is it already online in the library?

 

Looking forward to it!

 

Thanks

BR

[Clipper Ocean Spray 189]

New Rev. 1.2Q (beta) can be found here: 

 

http://www.mediafire.com/download/5kw1qanion0wqrf/AISIDSTAR12Q.zip

 

New:  AI will now speed up (except for final approach) as well as slow down to obtain traffic separation.  This should improve separation performance.  

 

New: AISIDSTAR now reads the wind direction at monitored airports for STAR assignment if other data is not immediatly available to determine actual landing runways in use.  The problem manifested itself at airports with little or no landing activity, where the program would assign STAR to the first arriving AI(s) that seemed to contradict actual landing conditions at the airport.  The program switches over to actual landing data once  it becomes available.

 

New: Sometime when the user creates a STAR with up-to-date NAV data, the runway numbers in the STAR and final approach files may not match the older runway numbers used by FSX.  This causes unpredicatable STAR and final selection.  For example, the user's STAR specifies runways 5 and 23, FSX however still uses the old runway 4 and 22 identifiers.  AISIDSTAR is unable to match runways 5 and 23 to the FSX runway causing the default STAR for runway 5 to be selected.  The new feature allows runway numbers to differing by one digit (e.g., runway 23 versus 22) to be selected if no exact matches were found.  AISIDSTAR will however generate a warning when this happens.  The user should double-check the STAR and final approach runway data.  If still no runway matches can be found, the pattern is selected having the closest entry point to the user's aircraft as before.

 

New:  AI on final approach will now skip first waypoint (IAF) and even second waypoint (1700 ft) if the program determines the AI needs to turn more than 90 degree to intercept the final.  This will prevent the AI from making sharp turns on final approach.  

 

New: User can now specify at what altitude (AGL) the AI on final approach switches to high-precision movement.  Above this altitude, the AI will behave much like a commercial jetline.  Below this altitude, the AI will turn much more quickly (if necessary) for demanding approaches (e.g., LOWI Rwy 8 Visual).  Add AIFinalPrecisionModeAlt (default is 2500 ft AGL) to the .ini.  You can also change the final precision turn scalar (AIFinalTurnScalar) and descent scalar (AIFinalAscentDescentScalar) to tweak further.

 

New: Banking animation improved when the user runs AISIDSTAR at a high sampling rate (AIHighSampleRate = 1).

 

New:  Added "extreme airports" package comprising Innsbruck (LOWI) and Toncontin (MHGT) to showcase the program's capability

to implement extreme final approaches.

 

New: Added a Quick Reference Guide that focuses on the user's goals and the parameters to change to accomplish those goals.

 

New: All AI target speeds (except ground roll) and manually entered waypoint speed restrictions are now Indicated Air Speed (IAS).

 

New: AI Monitor windows program now has a toggle button to include/exclude ground traffic.  Excluding ground traffic may significantly cut down on the size of the window making it easier to navigate.

 

New: Video documentation for version 1.2Q specifically (see below). 

 

Updated:  Autogen final approaches.  Revised code to better detect AI approaches that should be filtered out and not have their waypoint data recorded for subsequent autogen final approaches.

 

Updated: AI exiting STAR without timely landing clearance will no longer circle the airfield waiting for such clearance.  Instead, the AI lacking clearance will either be forced to land (autogen or custom final approach files exist) or will be vectored out of the area (autogen or custom final approach files do NOT exist).  I removed this feature because users did not realize the AI were circling the field.  Users instead were reporting to me the AI was turning back and forth in error (when it reality it was simply maintaining a circular track).  Circular holding (at stacked atltiudes) can however be turned back on (see the Quick

User Guide).  

 

Fixed:  The autogen final approach speed parameters (AIspeedIAF, AIspeed1700, AIspeed1000, AIspeed500, AIspeedtouchdown) were not getting applied when a stored autogen was read.  This has been fixed.

 

Important Revision:  A registered version (4.9x or later) of Peter Dowson's FSUIPC utility is recommended to use the full functionality of AISIDSTAR.  For example, the new AICull function requires FSUIPC.  For networked configuration users, Peter's WideFS (ver. 6.9) is also recommended.

 

Important New NOTAMS!

 

ENTERING FINAL APPROACH COORDINATES.  Do not use FSX when entering final approach coordinates for runway positions (e.g., touchdown, rollout) as the FSX coordinate display lacks the necessary precision (significant digits).  Use the freeware Airport Design Editor X (or similar) instead.

 

ENTERING RUNWAY DATA.   Ensure the runway number corresponds to the FSX runway description (for whatever airport scenery you are using).  Often, the latest FMS data will reference changed runway numbers that do not correspond to stock FSX runway descriptions.  For STAR files, if the runways do not match, AISIDSTAR will choose the STAR based on the waypoint having the closest entry point to the AI’s current position (i.e., not based on runway data), which often is satisfactory, but not always.  For Final Approach files, if the runways do not match, AISIDSTAR will not choose the final approach file having the incorrect runway data.

 

Please consult the following video documentation.

 

Video 1 - Working with the Included Sample SID/STAR Files and Creating New Files Using AIConv.exe:  http://www.mediafire.com/watch/mj3hgm33w24thcv/AISIDSTAR_Video_1_Sample_SID_STAR_Files_and_Creating_New_Ones_Using_AIConv.mp4

 

Video 2 - Custom Editing SID/STAR Files - http://www.mediafire.com/watch/t9irz80c497beo7/AISIDSTAR_Video_2_Custom_Edit_SID_STAR_Files.mp4

 

Video 3 - Operating Parameters.

 

What's New in Ver 1.2Q:  http://www.mediafire.com/watch/v2fgbtbu85h26xu/Whats_New_in_Ver_12Q.mp4

 

Part 1: http://www.mediafire.com/watch/urlrrdp09v5y5l4/AISIDSTAR_Video_3_Operating_Parameters_Part_1.mp4

Part 2: http://www.mediafire.com/watch/74e3kqyt4sndhwd/AISIDSTAR_Video_3_Operating_Parameters_Part_2.mp4

Part 3: http://www.mediafire.com/watch/24s4ek5x7xtiuf8/AISIDSTAR_Video_3_Operating_Parameters_Part_3.mp4

Part 4: http://www.mediafire.com/watch/24s4ek5x7xtiuf8/AISIDSTAR_Video_3_Operating_Parameters_Part_3.mp4

Part 5: http://www.mediafire.com/watch/24s4ek5x7xtiuf8/AISIDSTAR_Video_3_Operating_Parameters_Part_3.mp4

 

 

Thanks!

[Pete Dowson 558]

These links don't appear to be working for me. The others are okay.

 

Part 1: http://www.mediafire.com/watch/urlrrdp09v5y5l4/AISIDSTAR_Video_3_Operating_Parameters_Part_1.mp4

Part 2: http://www.mediafire.com/watch/74e3kqyt4sndhwd/AISIDSTAR_Video_3_Operating_Parameters_Part_2.mp4

Part 3: http://www.mediafire.com/watch/24s4ek5x7xtiuf8/AISIDSTAR_Video_3_Operating_Parameters_Part_3.mp4

Part 4: http://www.mediafire.com/watch/24s4ek5x7xtiuf8/AISIDSTAR_Video_3_Operating_Parameters_Part_3.mp4

Part 5: http://www.mediafire.com/watch/24s4ek5x7xtiuf8/AISIDSTAR_Video_3_Operating_Parameters_Part_3.mp4

 

 

Also, the last three appear to all named identically.

 

Best Regards

Pete

 

[Bryn 72]

Hi,

 

Really getting a lot out of this program - thanks very much.

 

What would be cool would be 'brake-to-exit' waypoints on the runway, that the aircraft decelerates to after landing.  If we added two (or more) of these waypoints, AISIDSTAR would randomly choose between them, and slow the aircraft to exit at the chosen waypoint.  This would help overcome the 135 degree turn-offs we often see when aircraft take opposite-end high-speed taxi turnoffs.  For airports without the waypoints, the AIRollOutSlowDownScalar would apply.

 

Bryn.

[Clipper Ocean Spray 189]

 

These links don't appear to be working for me. The others are okay.

 

 

 

Also, the last three appear to all named identically.

 

Best Regards

Pete

 

 

 

Pete, thanks for bringing this to my attention.  Here are the new links (I'll update the links in the readme as well).  These operating parameter videos are for an older version.  While still mostly relevant, I should probably redo them.

 

http://www.mediafire.com/watch/t7bhbs0x2gnrekn/AISIDSTAR_Video_3_Operating_Parameters_Part_1.mp4

http://www.mediafire.com/watch/t8nvlywnceoh67a/AISIDSTAR_Video_3_Operating_Parameters_Part_2.mp4

http://www.mediafire.com/watch/c3j1rmkvhoa6xas/AISIDSTAR_Video_3_Operating_Parameters_Part_3.mp4

http://www.mediafire.com/watch/t76l7wzds0zz9z8/AISIDSTAR_Video_3_Operating_Parameters_Part_4.mp4

http://www.mediafire.com/watch/24s4ek5x7xtiuf8/AISIDSTAR_Video_3_Operating_Parameters_Part_5.mp4

 

-Roland

[Clipper Ocean Spray 189]

Hi,

 

Really getting a lot out of this program - thanks very much.

 

What would be cool would be 'brake-to-exit' waypoints on the runway, that the aircraft decelerates to after landing.  If we added two (or more) of these waypoints, AISIDSTAR would randomly choose between them, and slow the aircraft to exit at the chosen waypoint.  This would help overcome the 135 degree turn-offs we often see when aircraft take opposite-end high-speed taxi turnoffs.  For airports without the waypoints, the AIRollOutSlowDownScalar would apply.

 

Bryn.

 

Thanks Bryn.  You can use the second waypoint to place on the centerline, but near an exit (autogen does this automatically).  The program is supposed to reduce the speed at a rate where the AI comes to a stop about 20 feet or so before the next waypoint it sees after touchdown (i.e., second waypoint), so the AI can exit the runway quickly.  I'm not sure how well that is happening though in reality, I probably need to look at it again for tweaking.  In the meantime, as you mention, you can play with the AIRollOutSlowDownScalar value.

 

Interesting idea about multiple brake-to-exit points.  The program is half-way there to doing this already.  If the program can't get the AI slowed down in time to the AISpeedEndRollout value for the next waypoint it sees after touchdown, it goes onto a third waypoint (if there is one) subject to the rollout timing out (AIRollOutTime).  I'll look into an option for making AIRollOutSlowDownScalar or AISpeedEndRollout random.

 

-Roland