PPRuNe Forums

PPRuNe Forums (https://www.pprune.org/)
-   Tech Log (https://www.pprune.org/tech-log-15/)
-   -   FSTD output for data collection (https://www.pprune.org/tech-log/635566-fstd-output-data-collection.html)

BoeingDriver99 17th September 2020 09:52

FSTD output for data collection
 
Hi there,

I was wondering what kind of data a modern FSTD Level D simulator creates. I'm attempting to do some research on pilot aptitude and wonder if sims produce raw data that could be analysed in some way. The devices must receive data from control inputs and then use that data to show the simulated aircraft behave in a corresponding way, also the motion system must use data from control inputs to simulate the feel of flying through the air/landing etc. I also recall the occasional print out of an approach or go-around etc which must be produced by said data.

Would anyone on here have in-depth technical knowledge that might help or maybe point me in the right direction towards someone who might know something?

Many thanks

Check Airman 17th September 2020 15:12

After doing the flight envelope upgrades to our sims (which I think are CAE), they printed out some fancy looking graphs at the end of the upset recovery training. Maybe contact CAE to get some technical details on that? I think that’s what you may be looking for.

BoeingDriver99 18th September 2020 03:05

I've emailed CAE and L3 and a few others to see if they can help.

Mad (Flt) Scientist 19th September 2020 02:46

If you speak to the right people you can get pretty much anything output from the sims, not just as plots but as actual data.That's what's used when a sim is used as an engineering test device for, for example, proving a windshear guidance system.

But you almost certainly need to be speaking with the actual sim design/build folks, not the folks running a training centre, since you'll need someone able and willing to "hack" into the raw data stream ("hack" in the sense that it's not the normal operations, rather than some kind of illegal operation :))

BoeingDriver99 19th September 2020 06:52

That is the kind of data I'd be interested in analysing for sure. So you reckon manufacturers rather than engineer/technicians who are maintaining the sims?

PEI_3721 19th September 2020 06:53

From dated experience in researching the mechanism and use of simulation in aviation, and then use of simulation involving pilot judgement and ability, I urge care in seeking guidance on pilot aptitude.
A danger is that with vast amount of data available from modern computers (simulators) you turn the soft people skills into numbers. You judge aptitude numerically opposed to qualitatively.

Too much focus on the data may result in a measure of an ability to fly 'a simulation' of a particular aircraft, in specific scenarios, and judging task execution - outcome, opposed to ability - process. Not what they achieved, but how they acted and their confidence in this and the future.
Situation choice is partially influential as this could represent the researchers biases and expectations more than rate the subjects character - a quality which can be related, but not measured.

'Select for Attitude, Train Skill' - aptitude ?

Consider the perspective of who is being rated and whom is judging (relate to 737 Max issues):-
https://www.newscientist.com/article...western-world/
https://www.newscientist.com/article...tern-cultures/

BoeingDriver99 20th September 2020 08:03

Thanks PEI, I understand the pitfalls you are talking about. I am currently just trying to see if the data is available and then see what kind of data it is and what could it be used for. From personal experience; instructors and examiners bring a lot of biases to simulator sessions and yes, ideally, they should be able to qualitatively and objectively assess trainees but anyone involved in flight training knows how varied individual examiners/instructors can be in their "objective assessment".

Further, with regard to recruitment, aviation is predominantly male and pale in the west and I'd like to learn more about how tests are or are not biased against certain demographics.

Your point about perspectives is crucial and they are interesting articles.

Thanks again

Mad (Flt) Scientist 30th September 2020 03:18


Originally Posted by BoeingDriver99 (Post 10888308)
That is the kind of data I'd be interested in analysing for sure. So you reckon manufacturers rather than engineer/technicians who are maintaining the sims?

Yes, if it's an option. They will be better placed to know what the "guts" of the device really contain, and what *could* be available, rather than perhaps having a more restricted list of what is routinely available/output.

They may also be able to warn you away from certain data - if, for example, certain aspects of the sim data are more effect-based than data based, then you might not want to be using those, and they will know that.


All times are GMT. The time now is 07:39.


Copyright © 2026 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.