Clipper Ocean Spray

Brian, Just checked KL71 with AIConv.exe and the file converted ok. Make sure you're using AIConv.exe version 1.4.11 dated 3/6/2015. Also, what is the file date of your KL71.txt file? The date I'm using is 1/6/15 (AIRAC Cycle : 1501) (navigraph). -Roland

Thanks Matias, much appreciated! I'll package it up with revision B and include instructions that the latest revision of AIConv.exe is included in the AIController main installation directory. -Roland

Thanks, a new version 1.4 with significantly more capability will be available any day now, so if you're checking the program out for the first time, it might be worth it to hold off for a few days. Regards, -Roland

Yes, the KDCA runway 19 river visual approach for AI is included in that package, as well as some other special approaches and departures associated with this airport. Latest version is 1.3B. -Roland

Hi everyone, I've just made available the latest beta version 1.3A, which now support AI jet routes and includes Matias's SID/STAR conversion utility. The program now controls the AI through all phases of flight from takeoff, SIDs, jet routes (new), STARs, final approach and landing. The old name (AISIDSTAR) gave a limiting impression of the program's capabilities. Thus, I've renamed my AI control utility from AISIDSTAR to AIController. I've started a new beta thread under the new name to avoid confusion: http://forum.avsim.net/topic/447974-ai-controller-sid-jet-route-star-and-final-approach-beta/ The old thread was very long too. Thanks everyone for your help and I hope you give a look and continue sending me excellent feedback and suggestions. The final released "old" AISIDSTAR version 1.2QR19 can be found in the avsim library under the file name: aisidstar12.zip -Roland

Abraham, The SIDs are selected based on the weighting the distance (including turn angle) from the AI to the SID entry point and from the SID exit point to the destination. So, a SID that has an entry point close and straight ahead of the AI and an exit that's "close" to the AI's destination may get selected even the the middle waypoints in the SID take the AI on a circuitous path. I would just remove some of the SIDs you don't like (de-activate). This is a problem when generating SID/STARs from nav data, sometimes to just get too many of them. You mention that you have 8 active SIDs now for that runway...I would deactivate about 4 or 5 of them (you can copy the files you don't want to used into a subfolder of the SID directory called "inactive" or the like). AICircleUntilAssignedRunway = 3000, AICirclingTimeNoRunway = 120, and AIHoldForRunwayPattern = 1 will cause AI that has exited a STAR, but not been assigned a landing runway by FSX, to begin a circular holding pattern in an attempt to eventually get a landing assignment. The circular pattern seems to help the AI's chances of getting assigned a landing runway. If you don't like the circular pattern, just remove those options and the AI will perform a downwind, base and final pattern after STAR exit. Yikes, that a huge separation distance, I would recommend =2 (the program is pretty strict about the separation distance in holdings modes 2-4 (AIholdingallowed) when the AI(s) join final, so a 5 nm separation distance will cut down landing rates dramatically. I actually think a 1 nm separation works pretty well. Yep, one video is worth a million words....oh well...but thanks for looking at it. I wish VoxATC had an option to turn off their airborne traffic (or at least the arriving VFR traffic lacking departure/destination data), but an option like that may be difficult and time-consuming to implement. -Roland

Nice, I haven't thought of that combination! Glad to see people experimenting with the options. There's several way to address the issue. You could add AIMaxSpawnDistanceFromAirport=30 to your .ini file. That will prevent any arriving AI spawned within 30 nm of the destination airport from being assigned a STAR. Also, if you're using the SIDSTARConverter program to creates STARs, make sure the "Create Short STAR" box has the number 30 or lower in it. You can also edit the STAR files yourself to achieve the same result. For example, if one STAR has five waypoints, where the entry waypoint is 40 nm out, the second waypoint is 30 nm out, etc., you can create two (or more) STAR files having the same basic STAR, but different entry points. E.g., the first STAR has the normal entry waypoint 40 nm out, the second STAR has the second waypoint as the first (entry) waypoint 30 nm out. Interesting idea, I'll look into it! One problem might be that many STAR(s) only have altitude ranges associated with the waypoint. -Roland

Bernd, AISIDSTAR uses its own configuration data and makes no permanent changes to FSX/P3D. AISIDSTAR is an external program that interfaces through the FSX simconnect and FSUIPC interfaces. If you don't run AISIDSTAR, FSX won't know it ever existed (there will be no trace of AISIDSTAR in FSX). The default AISIDSTAR.ini file only covers a few metro areas in the U.S., not the "whole North American area." Certainly, you can add SID/STAR files to cover most airports in North America, but that is something you have not done, correct? The default AI doesn't use navdata, the flights are direct. There are a lot of reasons why your AI might be missing. Have you downloaded and accidentally installed FS9 .bgl traffic files associated with North American airlines, thus disabling FSX airline traffic in North America? Did you check your recycle bin for deleted FSX files (e.g., *.bgl)? Have you installed/removed any traffic add-on packages recently? Sometimes that can cause problems, like leaving the "SimObjectPaths.X=SimObjects\Airplanes\AI Airplanes" statement out of your FSX config file. -Roland

Enzzo, can you email your custom STAR to me at my support address listed in the readme and I'll give it a look. Ahh...this is a separate issue from Matias' utility, but this is also a very good suggestion that's adds flexibility (people can still start AISIDSTAR after FSX the old way if they want to) and is easy for me to implement. Thanks for the idea (and to Onyx too). This would be especially useful to me (and others with a similar config) who run a networked laptop for Simconnect and Wideclient programs and don't mind letting AISIDSTAR "run" waiting for FSX/P3D to get loaded on the main machine (similarly to how I keep ActiveSkyNext running all the time). I think it would be useful for those running all programs on one machine as well. Good to hear, are you using the "AIImmediateForceLandAll" option? I actually have a new feature in the works...automatic jet route vectoring for ALL enroute AI (not just monitored airports) based on shortest path calculations from departure to destination. The jet route vectoring is basically working right now on my most current build, but I was thinking about adding an option where the user could enter average flight level wind information and the AI jet routes are chosen for shortest path taking into account the average winds. Do you (or anyone) think average winds is a useful option? -Roland

Enzzo, there are several options. The easiest is probably just add a single, additional exit waypoint to the STARs that generally places the AI on the side of the field that would have the left traffic pattern. That will trigger the program to execute a left pattern. Another option (more complicated but gives you more control over the pattern), add multiple custom waypoints to the STARs (with altitude restrictions) that extends the STAR into a left traffic pattern up to the endpoint of the downwind. After the AI exits this "extended" STAR at the end of the downwind, the AI will turn to base and final by themselves thus completing the left pattern. Another option, disable the traffic pattern and just have the AI directly intercept the IAF after exiting the STAR (add AITrafficPattern=0 to your .ini), in which case the AI will maintain a minimum AGL altitude (AIpatternalt in the .ini). Thank Matias, very nicely thought out utility! -Roland

Matias or Graham, Could you give the AICull feature a try and see how that works? Add AICull=2 and AICullHeight=18000 to the .ini. That should delete all airborne VFR traffic below 18000 feet near all airports monitored by AISIDSTAR. The best case result would be VoxATC detects the VFR traffic is no longer present and proceeds normally. -Roland

Yes, the AI should not have landed without clearance. Can you email me the aisidstar.ini file you're using? Maybe its a combination of different parameters causing unexpected results. Also, if you get a chance, add debugatcid1=TF-AMT to your .ini file (this is a new option I added to selectively debug specific AI aircraft). Then, zip up the aisidstar.log file and email it to my support address listed in the readme. Thanks! RC4 works fine with AISIDSTAR from personal experience. I think Pro-ATC/X and Pro Flight Emulator will work as well. Your aircraft and the AI traffic should follow the same assigned SID or STAR if both you and the AI are approach the STAR entry point from a similar position and landing on the same runway(s) covered by the STAR. However, if there are dozens and dozens of STARs active for a monitored airport in AISIDSTAR, the chances go down that you will end up on the same STAR as the bulk of the AI. One solution is deactivate excessive STAR files by moving them into a backup (inactive) directory (e.g., aidstar12/Chicago Metro Area Airports/KORD/inactive). Regarding SID(s), you're more likely to end up sharing the same SID with an AI because only a few usually correspond to a particular takeoff runway. Yes, there was some discussion of this before. If I remember correctly, departing AI work fine (when assigned SIDs at departure by AISIDSTAR). You can use AISIDSTAR for SID(s) only. As for arriving AI, the problem is VoxATC injects VFR AI for a circling approach with no departure or destination data, so AISIDSTAR doesn't control them (can't determine whether they're departing from and arriving at a monitored airport). It might be ok just to let VoxATC do its own thing regarding these arriving VFR aircraft. You could also try AISIDSTAR's AICull traffic density options (see the quick reference guide, p. 2 or the readme, p. 13), which delete AI not found to be departing from or arriving at monitored airports (i.e., the VFR VoxATC AI with no departure or destination data). However, I'm not sure what effect this would have on VoxATC (e.g., would it still think the deleted VFR AI are there?). -Roland

The latest AI monitor is expecting data (aitelemetry.txt) in a slightly different format. Make sure to replace the aimonitor.exe, aisidstar.exe and aitelemetry.txt files with the latest versions. A good approach might be to unzip the latest version into its own directory, test it to see if it the vanilla install works, then copy over your custom files/folders from your old installation. -Roland

AISIDSTAR has been uploaded to the AVSIM library and should now be available (aisidstar12.zip). Thanks everyone for your feedback and suggestions! Revised: Improved AI separation, so much so that you will have to adjust downwards the separation numbers you're used to entering. For example, AIseparationdistance=2 (nm) is now the "normal" separation number. To squeeze the AI tightly together on approach, I now recommend setting AIseparationdistance=1 (nm) in your AISIDSTAR.ini file. On the other hand, AIseparationdistance=3 (nm) now represents conservative spacing virtually eliminating all go-arounds regardless of traffic density. Note: Holding modes 2,3, or 4 must be active to see the enhanced holding (AIholdingallowed=2,3 or 4). New: If the user only wants AI with “official” FSX clearance to land, add AILandWithoutClearance = 0 to your AISIDSTAR.ini file. Background: AISIDSTAR will land any AI that fails to receive timely landing clearance provided an autogen or custom final approach file exists, however such AI must be deleted after the landing rollout because any AI that lands without clearance receives no taxi service to the gate. This new option avoids this problem. Fixed: Not establishing FSUIPC link in Prepar3D v 2.x. Various other fixes/revisions were also made. 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 in-game coordinate display (shift-Z) 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 might choose the final approach file having the incorrect runway data. Please consult the following video documentation: Overview of What's New in Version 1.2Q: Thanks! -Roland

aicullv11.zip will do that for you in the FSX utilities library. -Roland

Anthony, I just downloaded from the link and the AI Monitor is the latest version with the "Exclude Ground AI" button. Are you sure you haven't mixed versions? Sure, I'm interested, I have an "extreme approach" folder I'm including with the program. I add your approaches to the collection. -Roland

New Rev. 1.2Q Release 18 (beta) can be found here: http://www.mediafire.com/download/zs345x2pubm081q/AISIDSTARV12QR18.zip New: Regarding the new traffic pattern feature after the AI exits a STAR, I added options to specify the approach angle at which straight-ins occurs rather than a rectangular (crosswind, downwind, base) pattern (e.g., AIStraightInAngle = 45). I also added an option to turn off traffic patterns entirely (e.g., AITrafficPattern = 0). New: AIMonitor window more expandable vertically and horizontally so that user can see more data (if they want to use their display screen real estate to do so). Note: the column widths can be readjusted manually to suit a larger size. Fixed: Traffic pattern feature sometimes causing strange approaches and landings from weird angles. Revised: Turning off AISIDSTAR conducted go-arounds (AIGoAroundLeaveArea = 0) will now turn off short-final go-arounds. Please note however, if short-final go-around is turned off, AISIDSTAR will land every AI that makes it to "final approach" status regardless of late-breaking separation conflicts (some may like this however - I do - I prefer the AI to land even if a few must get squeezed in). Also, there is another class of go-arounds that AISIDSTAR must always perform regardless of whether AIGoAroundLeaveArea = 0. For example, if the AI exits the STAR but doesn't have a runway assignment after a time-out period elapses, then AISIDSTAR must perform a go-around otherwise under the control of FSX the AI would meander aimlessly around in or near the terminal area. 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 in-game coordinate display (shift-Z) 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 might choose the final approach file having the incorrect runway data. Please consult the following video documentation: Demo Video: https://www.youtube.com/watch?v=MMsO_FbrEf8 Overview of What's New in Version 1.2Q: http://www.mediafire.com/watch/v2fgbtbu85h26xu/Whats_New_in_Ver_12Q.mp4 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. Part 1: http://www.mediafire.com/watch/urlrrdp09v5y5l4/AISIDSTAR_Video_3_Operating_Parameters_Part_1.mp4 Part 2: http://www.mediafire.com/watch/t7bhbs0x2gnrekn/AISIDSTAR_Video_3_Operating_Parameters_Part_1.mp4 Part 3: http://www.mediafire.com/watch/24s4ek5x7xtiuf8/AISIDSTAR_Video_3_Operating_Parameters_Part_3.mp4 Part 4: http://www.mediafire.com/watch/t76l7wzds0zz9z8/AISIDSTAR_Video_3_Operating_Parameters_Part_4.mp4 Part 5: http://www.mediafire.com/watch/24s4ek5x7xtiuf8/AISIDSTAR_Video_3_Operating_Parameters_Part_3.mp4 Thanks! -Roland

If anyone gives it a try and has problems, send me a note and I'll give it a try. Sure, I've made the changes. The new version should be available shortly. Thanks for confirming Bluemja2002's issue. I think I now have this fixed in the new version. This is a very good suggestion, but I'm going to save it for when I look at taxi features. In the interim, have you looked at Ultimate Traffic 2? That program is very aggressive about re-injecting deleted AI, so much that it lessens the effectiveness of my AI Cull utility. -Roland

Just to verify, when you're spotting, do you have an overlay window up (e.g., a "new" window to view AI aircraft)? If so, a Simconnect bug causes bad data to be sent via the SimConnect link. AISIDSTAR then controls the AI aircraft erratically. It is OK to view map or another aircraft in the second window briefly, but if you keep the second window on indefinitely, AISIDSTAR will experience problems. Also, delete your KLAX final approach autogen files from the autogen directory. Sometimes a bad autogen slips through the cracks and gets recorded. Sure, I'll look into it. This is a popular request, although I might save it for the next version, where it would dovetail nicely with the new taxi parameters. Yes, it should. Are you seeing some problems? -Roland

Yes, basically the angle of approach to the airport in relation to the detected, landing runway. So, custom "pattern" waypoints are not needed, although they may help in certain situations if you're trying to do something specific. AISIDSTAR doesn't provide taxi services for now, although I'm looking at this feature for the next, major release, in which case I would definitely provide AI taxi speed and time-out parameters among many others. Regards, -Roland

New Rev. 1.2Q (beta) can be found here: http://www.mediafire..._in_Ver_12Q.mp4 Video 1 - Working with the Included Sample SID/STAR Files and Creating New Files Using AIConv.exe: http://www.mediafire...sing_AIConv.mp4 Video 2 - Custom Editing SID/STAR Files: http://www.mediafire..._STAR_Files.mp4 Video 3 - Operating Parameters. Part 1: http://www.mediafire...ters_Part_1.mp4 Part 2: http://www.mediafire...ters_Part_2.mp4 Part 3: http://www.mediafire...ters_Part_3.mp4 Part 4: http://www.mediafire...ters_Part_4.mp4 Part 5: http://www.mediafire...ters_Part_5.mp4 Thanks! -Roland

Thanks Warren, good to hear. Yes, when the AI Short Final Release option is active and the AI are released back to control of FSX at about 1000 ft, FSX is not able to execute extreme approaches (e.g., river visual, sharp turn at the bridge). Only AISIDSTAR can do that. However, I'll look into allowing lower, custom release altitudes. Perhaps a release at 400 or 500 ft AGL might actually work. Yes, I won't be able to do that for version 1.2, which I want to release. However, I'm looking to add it for the next big version. I'm actually leaning toward replacing the taxi system, but you're right, the .bgl reading and sorting will be the most difficult part, harder than taxi routing. I also wanted to mention I'm planning to get a developer's license for P3D, I'll email them and ask if they can add a Simconnect feature that releases the AI for taxi once it lands. Thanks, I think I have software that sets landing lights, but I honestly don't remember the product! The first line, just enter the ITZ02 waypoint, then a dummy waypoint somewhere along the heading 141 path. The program doesn't process "VECTORS", you could just end the SID file at that point so FSX can take over and start vectoring the AI. For the second through fourth lines, the AIConv utility wouldn't convert these lines. You would have to go custom. For example, enter a dummy waypoint at heading 151 where you anticipate the AI to acheive 650 ft, then a second dummy waypoint somwhere along the heading 201 path, then end the SID there for FSX vectors. You would also need to add AIClimbOutHeight = 650 to your .ini file because default is 900 ft. -Roland

Here a Youtube demo video I made: https://www.youtube.com/watch?v=MMsO_FbrEf8#t=18 I'll probably reference the video in the docs. What do you guys think? FRAP videos of FSX aren not my strength. Hopefully it demonstrates interesting capabilities of the program without the video quality being too distracting. -Roland

Leo, got your email...please check for response. Thanks, I'll look into it...this might not be feasible with UT2, but might work with all the .bgl based traffic addons. Yes, the animation is more primitive on the SIDs. I spent a lot of time getting the AI to look right on final approach, but less time for departure. I test at KLAX all the time with crazy AI densities (UT2 = 100% and WorldofAI = 100% at the same time). Make these changes to the default .ini files: AIseparationdistance=5.0 (or AIseparationdistance=3.0 for tighter spacing) UserAIseparationdistance=3.0 AIholdingallowed=3 AIVectorSortPeriod = 1 (if running AISIDSTAR on a separate client machine). If UT2 make the following changes: UT2Compatibility=1 AImaxspawndistancefromstar=70 -Roland Warren (or anyone else having this problem), any feedback on this test version: http://www.mediafire...RFLOWN_TEST.zip ? -Roland

Do you use the AIShortFinalRelease = 1 parameter in your .ini file? If so, there might be a bug when that parameter is used that could cause this problem, please give this test version a try: http://www.mediafire.com/download/03770han1jtb4d1/AISIDSTAR_OVERFLOWN_TEST.zip Also, add AIFinalApproachTooLow = 0 to the .ini as a separate trouble-shooting step and see if that makes any difference. Are you using a registered version of FSUIPC and the AIShortFinalRelease = 1 parameter? If so, you might be seeing the same bug Warren reported above. Please give this test version a try: http://www.mediafire.com/download/03770han1jtb4d1/AISIDSTAR_OVERFLOWN_TEST.zip