AI Aircraft SID and STAR Controller

[Victory103 2]

Not a PMDG flyer yet, but do have some payware add-on gauges from ISG that includes a working FMS. I'm trying to use AIconv.exe vice having to hand jam all the SID/STARs, the ISG contains a Navdata folder with sid and start .txt files, albeit not the current AIRAC. I just need to see what format the navdata for PMDG looks like. AIconv is getting the info, but does not complete the conversion.

[ysfsim 22]

Not a PMDG flyer yet, but do have some payware add-on gauges from ISG that includes a working FMS. I'm trying to use AIconv.exe vice having to hand jam all the SID/STARs, the ISG contains a Navdata folder with sid and start .txt files, albeit not the current AIRAC. I just need to see what format the navdata for PMDG looks like. AIconv is getting the info, but does not complete the conversion.

if you purchased an airac cycle from navigraph, you get access to all the available brands.

[Clipper Ocean Spray 189]

I would really like the air cull function to work as I had suggested in a separate program. Us that posdiblw , Roland?

 

 

Sure, here it is, a stand-alone AICull utility (intelligent AI traffic density):  http://www.mediafire.com/download/0jph897gu7asdu7/AICULLV10.zip

 

Can you give a try (and anyone else) and let me know if there are problems?  It actually worked out well breaking the AICull function into a separate program.  Thanks for the suggestion.  You can also update the list of monitored ICAO(s) after the program starts, as you suggested.  

 

 

I was in Boston last night and less than say 10% of the air land after a long time. I think the 10 mile radius for ai circle until assigned approach needs to be shrunk?

 

 

Add AIHoldForRunwayPattern=2 to your .ini.  The AI will circle in a decreasing radius "hunting" for the approach assignment.  

 

To me personally it doesn't seem worth it making the AI hold if it doesn't get a runway assignment right away.  If AISIDSTAR forces an AI to land because it doesn't get an assignment, all that happens is the AI is deleted after finishing its rollout (because it won't have taxi service and will just sit there).  Better that than having another AI "holding."  Most AI will still get their assignments and land normally with taxi service to the gate.  You can turn circle holding (waiting for approach assignment) OFF by adding AICircleUntilAssignedRunway=0 or AICirclingTimeNoRunway=0 to your .ini.  

 

 

I alsowould llike to see the ability to create holding points that are runway specific to allow planes to circle instead of the airport. Theses points can be near the required runway to allow fsx to assign approach easier.

 

 

Those are good suggestions.  I've tried them, but they don't increase how quickly the AI are assigned approach clearance.  

 

-Roland

[ysfsim 22]

Thanks a lot Roland. This is going to be fun.

[ysfsim 22]

Those are good suggestions. I've tried them, but they don't increase how quickly the AI are assigned approach clearance.

What is this "CustomHoldWaypoints" file for?

[ysfsim 22]

okay, I tried the aicull program and it works..mostly.It misses some airports. I have it set to scan ground traffic and min altitude set at 1000ft.

[Clipper Ocean Spray 189]

What is this "CustomHoldWaypoints" file for?

 

 

A lot of approach files have published holding points, holding altitudes, and even circling directions.  If you want the AI to hold in a published manner at a waypoint, then add the waypoint to CustomHoldWaypoints along with the holding data.   E.g., ENI,R,2000 means AI will hold at ENI (if a separation conflict occurs) making a circling right turn at 12000 ft MSL. Trevor at FSCaptain did a really nice job with the included STAR/SID(s) in the SFO metropolitan area, including adding published hold data for waypoints.

 

-Roland

[ysfsim 22]

A lot of approach files have published holding points, holding altitudes, and even circling directions.  If you want the AI to hold in a published manner at a waypoint, then add the waypoint to CustomHoldWaypoints along with the holding data.   E.g., ENI,R,2000 means AI will hold at ENI (if a separation conflict occurs) making a circling right turn at 12000 ft MSL. Trevor at FSCaptain did a really nice job with the included STAR/SID(s) in the SFO metropolitan area, including adding published hold data for waypoints.

 

-Roland

I see thanks. Didn't know Trevor was involved.

[Clipper Ocean Spray 189]

okay, I tried the aicull program and it works..mostly.It misses some airports. I have it set to scan ground traffic and min altitude set at 1000ft.

 

Great, I like 10000 too (I want to keep the enroute traffic density normal).  If you see a few AI at an airport that shouldn't have any AI traffic, chances are those AI are departing from or arriving from an airport on the "do not delete AI" list.

 

Edit: Oops it was Travis (not Trevor) at FSCaptain.  All the files he wrote for the SFO metro area are custom.  Ronald also did a great job writing custom files for the LAX metro area.  Also, as I mentioned earlier, andqui gave me the suggestion for AICull (he basically wrote me an email telling me I already had written the code to do it, I just didn't know it yet).

 

-Roland 

[ysfsim 22]

Great, I like 10000 too (I want to keep the enroute traffic density normal). If you see a few AI at an airport that shouldn't have any AI traffic, chances are those AI are departing from or arriving from an airport on the "do not delete AI" list.

 

Edit: Oops it was Travis (not Trevor) at FSCaptain. All the files he wrote for the SFO metro area are custom. Ronald also did a great job writing custom files for the LAX metro area. Also, as I mentioned earlier, andqui gave me the suggestion for AICull (he basically wrote me an email telling me I already had written the code to do it, I just didn't know it yet).

 

-Roland

I understand but when I tested it, only lax and SFO were on the list. I saw planes in places like Klgb kbur ksna etc. When I flooded the region with game aircraft ( over 900 aircraft simulated rather than the average of 200 plus for the lax area) the program went to work and cut that list down but still left some again.

[ysfsim 22]

I understand but when I tested it, only lax and SFO were on the list. I saw planes in places like Klgb kbur ksna etc. When I flooded the region with game aircraft ( over 900 aircraft simulated rather than the average of 200 plus for the lax area) the program went to work and cut that list down but still left some again.

sorry I meant to say ga aircraft. My average for airliner traffic in the lax area is around 200.

[Clipper Ocean Spray 189]

sorry I meant to say ga aircraft. My average for airliner traffic in the lax area is around 200.

 

Thanks, I did the same (flooded the area traffic including GA).  AICull is working correctly on my end.  I positioned the user aircraft at KLAX.  After running AICull, I noticed traffic at KLGB, KBUR and KSNA, however this was commercial and/or GA traffic either departing from or arriving at the airports in the ICAOList (i.e., KLAX or KSFO).  I verified this using the FSX Traffic Tools Explorer.  

 

I have a question for you (or anyone else who cares to answer).  I'm close to uploading the final AISIDSTAR v1.2 into the library.  However, regarding autogen approaches, I have the default set so that AI "slows" down on short final at altitudes corresponding to typical flap deployment.  However, more than one user has reported to me that they actually thought the speed control on short final was to maintain traffic separation. 

 

Did anyone else think the program was changing the speed of AI on short final because of traffic separation?  The problem is, if that is the reason why a user thinks the AI on short final is slowing down, it creates a bad impression because the user would think the AI closest to landing (i.e., the AI at full flaps ready to land) should not be slowing down (rather the AI further behind should be slowing down).

 

I can set the default so that the AI perform the short final at one speed (e.g., 145 kts).  However, I'm pretty sure I'll then get feedback that this is unrealistic (even though I provide the parameters in the readme to change to a variable speed profile for final).

 

I think I'm in a catch-22 here, just curious what other people think.

 

-Roland 

[bib974 0]

Hi Roland,

 

Sorry i havent been able to revert yet but thank you for the time spent in your replies ever since.

I have continued tedting and tweaking your great program and actually started using autogen approach and design my own. Looks promising for now. One issue I had with autogen approach is the Ai plane flying right and left and right and let from glide slope intercept to touchdown. Dont know if it is a bad tweak on my side, will keep testing and look into it.

I have some other remarks i would like to share for.future updates as well, will write you an email.

 

For your question about autogen final approach speed, i did think tobhave read somewhere it was for separation purpose. Im in favor of slowing the ai from 160kias on long final to 140 before touch, although customizable speeds depending on aircraft category/type would be more realistic.

 

Thanks for all!

 

BR

[ysfsim 22]

I have noticed some patterns. Some airports are fine without approach files and some are not. It would seem that the busier airports should have an approach file to function properly. It is a pain to have to make the approaches manually so... could you provide some way for us to create autogen approaches for selected airports. It could be something like the aiconv.exe

 

thx

[bib974 0]

I noticed that sometimes autogen approach are saved in the autogen cached.folder in aisidstar folder. Might be a good start fir further.customization. Is it so intended Roland? However it seems that the file gets deleted once the program exits. I couldnt see the file at all times either when running the program.

Couldbyou shed some liggt on this?

Thank you!