New feature: Virtual pilot controls the take-off

[cavaricooper]

With PSX's ability to have 2 pilots on the deck, I keep wondering why we aren't looking at the IP/teacher functioning as the PF for these issues, with the student pilot able to go heads down, in real time? Also important is the skill to work through the lesson, AND LEARN the skills necessary to be functionally aware, using CRM techniques.

I am not a fan of unreal automation, but I understand this is only one opinion, and not necessarily the popular one. I also understand that in a classroom with multiple students, this might not be possible.

FWIW- C

[Avi]

It's interesting to think of how the aircraft would get from the gate to the runway. Probably go directly to the midpoint and then taxi along a downwind leg, base, and final. But what if the "gate" is close to the midpoint? Hmmm, gets complicated.

What if the gate is close to the runway (threshold)? Why to get to the runway midpoint and go back? 

[Will]

Avi,


To be honest, I wasn't thinking about getting the aircraft to the takeoff position in the most efficient and fastest way. I was thinking of a taxi experience that was somewhat standardized, and included some "downwind" parts and then two 90º turns to line up on the runway.


But there's no a priori reason for doing that. Perhaps the aircraft could taxi to whatever is closer, the midpoint of the active runway, or a spot 500 feet from the departure end of the runway. Or something else?

[Hardy Heinlin]

Hmmm, I'll need to update my additional pilot... I'll just record those de novo probably. Will you add the wav files as the last ones in the existing list?

My plan is to create those additional wav files for all default and add-on pilots, i.e. for Will and Sylle too.

"Checked" -- from "Check your rudder" and "Check your speed"
"Gear up" -- from "Positive rate, gear up" (if necessary)
"Your control" -- from "Check your rudder" and "Fuel control switches"


|-|ardy

[Hardy Heinlin]

I am not a fan of unreal automation, ...

This is not a systems automation, this is a simulation of a human pilot.

Just like all the other simulations of the other pilot's silent tasks, call-outs etc., and the tug driver, ATC, traffic, flight attendant ...

And nobody is forced to use those simulations. Just don't click it.

[asboyd]

Hardy, I think it is a great idea and will help a lot of younger trainees get to see how it feels to be the PNF and work at that level for experience. Also sometimes it is good to have a role played out so you can try other things without having to concentrate on the takeoff....

Cheers,
Alex B

[Hardy Heinlin]

Yes, Alex, I think so too. The simulation is now implemented and I must say it is -- surprisingly -- even more interesting than expected to watch the other pilot's reactions on extreme crosswinds and various combinations of engine failures before and after certain V speeds ...


Cheers,

|-|ardy

[andrej]

Dear Hardy,

from a non-pilot point of view, I believe that this is an interesting feature. It will at least show me, what is the proper procedure to take off. I have read the manuals, watch many videos, but this provide 'hands on' experience.

I concur that 'auto-taxi' feature would add value as it would show me how to properly taxi (especially onto a runway).

Nonetheless as others have stated, it is nice to do PNF duties as well. This is probably very next best thing next to a shared cockpit.

Thank you!

[cagarini]

From an AI perspective  ( my area before moving to meteorology... ) I am looking forward for that Virtual Pilot :-)

How did you "train" him ?

[tango4]

Hi Hardy,
As always,  thanks for keeping to add new features to this Sim. This virtual co-pilot for takeoff looks awesome.

I just have one question regarding this item in your first post:
"Virtual pilot will set idle thrust if any EICAS warning, caution, or advisory message occurs before V1 minus 5."
If I recall correctly many airlines split this phase in two.
Before a certain speed (usually 60 or 80 knots depending on aircraft and airline policy), the crew will abort for basically any abnormal as in your proposed feature. During the high speed phase the crew will only abort for a "major" issue (engine failure/fire, any master warning).
Perhaps I'm wrong or perhaps it is simply too complicated to implement.

Charles

[Hardy Heinlin]

I could make it ignore advisory messages when above 80 knots.

How's that? Advisories. Above 80.


|-|ardy

[tango4]

That sounds great !
I'd just wait for a real pilot's confirmation just in case if possible.

Charles

[emerydc8]

 Rejected takeoff. EICAS caution alerts include illumination of the Master warning/caution light. It is not necessary to reject for advisory messages. Advisory messages are EICAS alerts with no Master warning/caution light and no aural alert. Caution lights (of the Master warning/caution) are inhibited above 80 knots. Warning lights (of the Master warning/caution) are inhibited above V1. For windshear and for any fire during takeoff, the Master warning/caution is illuminated. Windshear caution alerts are inhibited above 80 knots. Windshear warning alerts are inhibited above 100 knots.

 Captain makes decision to reject. Reject briefing example: Prior to V1 we will reject for any Master warning/caution light, engine failure, or, the aircraft is unsafe or unable to fly. This brief encompasses a reject at
any speed due to Master warning/caution logic.

Below 80 knots:
 Clearly announce any abnormalities using the below 80 knots criteria.
 Master CAUTION lights are normally associated with major aircraft systems such as HYD PRESS SYS __.
Non-plug type doors (overhead hatch and all of the cargo doors) will produce a Master CAUTION light. Plug
type doors (for example L1 door) will not.
 It is not necessary to reject for EICAS advisory messages. Advisory messages are EICAS caution alerts with
no Master CAUTION light and no aural alert.
Above 80 knots:
 Clearly announce only any Master WARNING light, fires, fire warnings, engine failures, PWS warnings, or, if
the airplane is unsafe/unable to fly.
Captain: Below 80 knots reject for:
 Master WARNING or Master CAUTION light.
 System failures.
 Unusual noise or vibrations.
 Tire failure.
 Abnormally slow acceleration.
 Takeoff configuration warning.
 Engine failure.
 Fires.
 Fire warnings.
 PWS warning and caution alerts.
 The airplane is unsafe or unable to fly.
Captain: Above 80 knots and prior to V1, reject for:
 Master WARNING light.
 Engine failure.
 Fires.
 Fire warnings.
 PWS warning.
 The airplane is unsafe or unable to fly.

[tango4]

Many thanks, Jon, that clarifies perfectly.


Charles

[Hardy Heinlin]

Now available in 10.1.0-beta1:

http://aerowinx.com/board/index.php?topic=3858.0

[Jeroen Hoppenbrouwers]

Is the virtual pilot sneakily using the autopilot for her flying, or is it a new piece of software?

[Hardy Heinlin]

New piece of software.

[Will]

Now, to get even more meta, we need a simulated human user operating PSX. As in, pressing "Initiate the human initiating the takeoff," a new piece of software will open the instructor page by pressing Esc, then move the mouse to the "Initiate" button, and press it with a mouse click.

[cagarini]

I can't get the "Initiate" button to become available ?

Chose the St-Marteen situ and as  I go to Human, the button is there, in the lower right corner, but greyed out ?

Ddn't read the release notes :/  Sorry...

[Hardy Heinlin]

I will have to think about an algorithm for the virtual pilot's behaviour when the air data disagree.

There are four information sources for the human: Two EFIS airspeeds, one standby airspeed, and the airspeed impression by looking out the window.

E.g.

When the left and right airspeeds disagree when either one reaches 80, the VP will start RTO and won't call "Checked".

After 80, when the true pilot's airspeed goes to zero or doesn't rise, VP will start RTO if below V1, else continue using the VP's airspeed. If the VP's airspeed goes to zero or doesn't rise, the VP will use the standby airspeed. If that fails, the VP will rotate when the groundspeed is 200.

Aside from that, the EICAS message "IAS DISAGREE" will already have an effect in the existing, message dependent behaviour logic.


|-|ardy