Jump to content

autopilot function


Oldguy
 Share

Recommended Posts

I'm just running a simple, canned mission for the T model, to allow the plane to fly itself while I play around with control settings and the like, for my VR setup. It comes with the free version of DCS and involves a simple take-off, follow multiple way-points, and land.

 

I've got plenty of fuel, and time-over-wp is not important, so I take-off and keep the power high in navigation mode. I settle on a course close to what the FP requires, and I turn the AP on. I just want the plane to follow the mission (as written by someone on the DCS team), so I can screw around with other settings and enjoy the VR view.

 

Thing is, I thought the plane was supposed to follow the plan's way-points, including altitude. I can see the plan altitude and speed values in the HUD, but the plane rarely matches either. It does not even seem to try to climb to match the planned WP's altitude.

 

The distance-to-next-waypoint counter, at the bottom of the HUD, sometimes increments up, and occasionally down. The AP cycles thru the WPs correctly, by number, but the course the plane flies from one to the next seems very unpredictable.

 

Since this mission crosses some very high mountain peaks, it seems important that the autopilot try to keep the plotted altitude, or at least somewhat close to it.

 

So, in a round-about way, I'm asking if the DCSW autopilot is functioning normally, or is this unpredictable behavior the result of buggy behavior?

Link to comment
Share on other sites

I suggest to look into this thread - I haven't tested it in the last version, but altitude hold AP and its related modes work well only when the SL baro pressure on the map is set to 760. Basically the altitude hold AP does not use the reading of the on-board altimeter but uses some own altimeter set always to 760 (I think it is a relict of old modeling when we weren't able to change QNH of the altimeters in FC1):

 

https://forums.eagle.ru/showthread.php?t=150622&highlight=Su-25T+autopilot+bug

Link to comment
Share on other sites

I'm just running a simple, canned mission for the T model, to allow the plane to fly itself while I play around with control settings and the like, for my VR setup. It comes with the free version of DCS and involves a simple take-off, follow multiple way-points, and land.

 

I've got plenty of fuel, and time-over-wp is not important, so I take-off and keep the power high in navigation mode. I settle on a course close to what the FP requires, and I turn the AP on. I just want the plane to follow the mission (as written by someone on the DCS team), so I can screw around with other settings and enjoy the VR view.

 

Thing is, I thought the plane was supposed to follow the plan's way-points, including altitude. I can see the plan altitude and speed values in the HUD, but the plane rarely matches either. It does not even seem to try to climb to match the planned WP's altitude.

 

The distance-to-next-waypoint counter, at the bottom of the HUD, sometimes increments up, and occasionally down. The AP cycles thru the WPs correctly, by number, but the course the plane flies from one to the next seems very unpredictable.

 

Since this mission crosses some very high mountain peaks, it seems important that the autopilot try to keep the plotted altitude, or at least somewhat close to it.

 

So, in a round-about way, I'm asking if the DCSW autopilot is functioning normally, or is this unpredictable behavior the result of buggy behavior?

It's seldom that the in-cockpit vs mission editor (and consequent required altitude) are identical. I've forgotten exactly why.

 

As far as your odd course WP to WP is concerned. The system is attempting to regain the courseline between the two WPs. It continues intersecting it in increasingly smaller increments until it's on the true line. If the turn requires a small course change, it's not a big deal. If it's a large turn, however, acquiring the course line can take awhile.

 

BTW, LAlt - ~ (press and hold) used to allow you to take temporary control of the aircraft and adjust your altitude in Enroute Mode. When you released that key combo, the AP followed the course but at the new altitude. Don't know if that's still the case, though. It's been awhile and I'm not where I can check.

YouTube Channel: https://www.youtube.com/channel/UCU1...CR6IZ7crfdZxDg

 

_____

Win 10 Pro x64, ASUS Z97 Pro MoBo, Intel i7-4790K, EVGA GTX 970 4GB, HyperX Savage 16GB, Samsung 850 EVO 250 GB SSD, 2x Seagate Hybrid Drive 2TB Raid 0.

Link to comment
Share on other sites

Could you save a track and attach it ?

 

Here you go.

 

I do not know what camera the track file uses, but of course, it does not match the PoV of my HMD while in-game, so I can only say that what I saw was remarkably smoother and more stable when I made the track.

 

The best way to try to decipher this is to watch the way-point number in the HUD, and the distance counter just to its left.

 

If the aircraft passes the way-point, and the distance-to counter starts to rise, then I manually shift to the next WP.

 

The AP begins to act a bit wonky beginning with WP 3, which I went past by quite a bit before I manually shifted.

 

Thereafter, it will occasionally increment up to the next WP automatically, but that is the exception, not the rule.

 

The view in the mirrors is funny, just as an aside: you get a nice look down into the bowels of the A/C, with the spinning disc representing the intake fan blades.

 

On the occasion when it appears that I will run aground, I take manual control and try to stay close to path until it is safe to shift back to AP.

 

Otherwise, this track is pretty indicative of every trial I have made with this canned mission.

 

Any ideas?

 

:joystick:

autopilot test.zip

Link to comment
Share on other sites

Here you go.

 

I do not know what camera the track file uses, but of course, it does not match the PoV of my HMD while in-game, so I can only say that what I saw was remarkably smoother and more stable when I made the track.

 

The best way to try to decipher this is to watch the way-point number in the HUD, and the distance counter just to its left.

 

If the aircraft passes the way-point, and the distance-to counter starts to rise, then I manually shift to the next WP.

 

The AP begins to act a bit wonky beginning with WP 3, which I went past by quite a bit before I manually shifted.

 

Thereafter, it will occasionally increment up to the next WP automatically, but that is the exception, not the rule.

 

The view in the mirrors is funny, just as an aside: you get a nice look down into the bowels of the A/C, with the spinning disc representing the intake fan blades.

 

On the occasion when it appears that I will run aground, I take manual control and try to stay close to path until it is safe to shift back to AP.

 

Otherwise, this track is pretty indicative of every trial I have made with this canned mission.

 

Any ideas?

 

:joystick:

Assuming you've been following the same startup proceedure for your flights, your error occurs in the first 3 minutes after entering the cockpit. I hadn't been paying attention during startup but, as I watched you take the runway and turn on NAV mode, I noticed the HSI compass card jump. Told me all I needed to know. I went back to doublecheck and...

 

Afterturning on electrical power, it takes 3 minutes for the HSI gyros to align before you can move the aircraft. You were about 20 seconds or so short. From that point on you were guaranteed to have issues. EDIT: That little tidbit is not, AFAIK, in the manual. It's only mentioned in one of the training tutorials. It would have been beneficial to have it reinforced in this tutorial, too. :EDIT ENDS

 

Attached is a track of my own made from your track. I took control after you entered the cockpit. I didn't bother to close out and start up TIR, so the views are kind of boring but you'll see the AP follow the WPs perfectly. I quit shortly after WP5 or so.

 

The banking left and right after the turn is the system attempting to align the courseline and WP needles. That's the purpose of the long focus on the HSI after one turn where I knew there'd be enough time for the AP to complete the job before the next WP.

 

Also, I wasn't seeing any large deviations in altitude (required vs obtained) The only substantial "difference" in numbers I saw was when the the radar altimeter was giving you the reading rather than the barometric.

 

BTW, I did not view your complete track. I stopped more or less where I did in mine. But, by then, I figured I knew what the root cause of your issue was.

Ironhand_autopilot test.trk


Edited by Ironhand

YouTube Channel: https://www.youtube.com/channel/UCU1...CR6IZ7crfdZxDg

 

_____

Win 10 Pro x64, ASUS Z97 Pro MoBo, Intel i7-4790K, EVGA GTX 970 4GB, HyperX Savage 16GB, Samsung 850 EVO 250 GB SSD, 2x Seagate Hybrid Drive 2TB Raid 0.

Link to comment
Share on other sites

Okay, I'm going to give the three minute drill a shot, and see if that prevents the AP from overrunning the WPs repeatedly; I had to manually advance them once the AP overshot.

 

This time, I will simply watch the run.

 

Thanks for the tip; much appreciated.

Link to comment
Share on other sites

Okay, I'm going to give the three minute drill a shot, and see if that prevents the AP from overrunning the WPs repeatedly; I had to manually advance them once the AP overshot.

 

This time, I will simply watch the run.

 

Thanks for the tip; much appreciated.

 

EDIT:

 

OK, ran it again, and in an excess of zeal, waited probably 4 1/2 minutes after power-on, before selecting navigation mode at hardstand, then taxiing for takeoff, selecting AP after climb-out and reaching WP 1 altitude.

 

Once again, it did not increment up to WP 4 after overrunning WP three.

 

Track is attached.

 

Many thanks for your patience and sharing your experience. I'm going to run your track while I wait on this reply.

autopilot test 2.zip

Link to comment
Share on other sites

Unfortunately it'll be tomorrow morning before I can view your 2nd track. FWIW, in mine, the WPs advance as they should. In your first track you were carrying a lot of speed between WPs 2 and 3 which are quite close together with a fairly sharp turn. I wonder if the speed caused you to overshoot. A possibility, I suppose. Anyway, I'll check your 2nd track in the morning when I'm in front of my computer again.

YouTube Channel: https://www.youtube.com/channel/UCU1...CR6IZ7crfdZxDg

 

_____

Win 10 Pro x64, ASUS Z97 Pro MoBo, Intel i7-4790K, EVGA GTX 970 4GB, HyperX Savage 16GB, Samsung 850 EVO 250 GB SSD, 2x Seagate Hybrid Drive 2TB Raid 0.

Link to comment
Share on other sites

Unfortunately it'll be tomorrow morning before I can view your 2nd track. FWIW, in mine, the WPs advance as they should. In your first track you were carrying a lot of speed between WPs 2 and 3 which are quite close together with a fairly sharp turn. I wonder if the speed caused you to overshoot. A possibility, I suppose. Anyway, I'll check your 2nd track in the morning when I'm in front of my computer again.

 

Yes, that is possible, I suppose. I would think it would only affect ToW, so I deliberately kept speed high, so the little guy wouldn't drop out of the sky, as it had done on earlier runs, when the AP made sharp turns and power was in the 80% range.

 

Watching your track, it does indeed advance correctly, so I will experiment with speed, and try to hit 3 at a lower velocity.

 

Thanks for your continued patience.

 

Post script:

 

In fact, that seemed to help, although I saw AP turns above 700 knots; I do not know what Max speed + Max angle limits are. The sharpness of turn and distance to WP are definitely a factor.

 

In any case, the mission designer had a real sense of humor, with altitude set at 2000 meters in the highest mountains.

 

Made it to Nalchik, nonetheless. I'll just have to play around with the AP and find out what the combination limits are.

 

Thanks again for your help!


Edited by Oldguy
Link to comment
Share on other sites

...

Post script:

 

In fact, that seemed to help, although I saw AP turns above 700 knots; I do not know what Max speed + Max angle limits are. The sharpness of turn and distance to WP are definitely a factor.

 

In any case, the mission designer had a real sense of humor, with altitude set at 2000 meters in the highest mountains.

 

Made it to Nalchik, nonetheless. I'll just have to play around with the AP and find out what the combination limits are.

 

Thanks again for your help!

Your airspeed's only a factor, when the WPs are set close together with a fairly sharp turn. In your 2nd track, the closest you got to the WP on the initial turn was 3+ km vs around 1 k, when I took it slower.

 

Have fun experimenting. Just make sure that, when you see the WP advance to the next WP, if there's a climb, you immediately advance the throttles. In this mission, you're never so heavy that that approach won't get you to the top OK.

YouTube Channel: https://www.youtube.com/channel/UCU1...CR6IZ7crfdZxDg

 

_____

Win 10 Pro x64, ASUS Z97 Pro MoBo, Intel i7-4790K, EVGA GTX 970 4GB, HyperX Savage 16GB, Samsung 850 EVO 250 GB SSD, 2x Seagate Hybrid Drive 2TB Raid 0.

Link to comment
Share on other sites

  • 1 month later...

Which version of Su-25 did you fly? I didn't find auto pilot.


Edited by mytai01

MS Win7 Pro x64, Intel i7-6700K 4.0Ghz, Corsair RAM 16Gb,EVGA GeForce GTX 1080 FTW GAMING ACX 3.0, w/ Adjustable RGB LED Graphics Card 08G-P4-6286-KR, Creative Labs SB X-FI Titanium Fatal1ty Champ PCIe Sound Card, Corsair Neutron XTI 1TB SSD, TM Warthog Throttle & Stick, TM TPR Pedels, Oculus Rift VR Headset CV1, Klipsch Promedia 4.1 Speakers...

Link to comment
Share on other sites

 Share

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...