search

fgx / fgx.github.io

The FGx public web presence on GitHub
http://fgx.github.io
Other
2 stars 3 forks source link

Update ~ 2016-04-15 ~ Fgx FlightPath R2 #1

Closed theo-armour closed 8 years ago

theo-armour commented 8 years ago

image

FGx FlightPath R2

@geoffmcl @pedromorgan

I'd like to start chatting using GitHub issues. Is this OK with you?

GitHub issues have tags and can be closed and use Markdown and do all sorts of other cool things

🔗 FGx FlightPath

Features

Keeping the same naming convention for time being on my side because I think it's nice for a newcomers to be able to look at a title and get an idea as to what the script does.

Feel free to fork and rename. ;-)

Remarks

Can we talk about what would be the ideal data set?

To me, it might be

julian date, latitude, longitude, altitude, roll, pitch, yaw followed by details such as landing gear, flaps and any text notes as well.

Captured every second or sub-second? ASCII or Binary?

Does Flightgear have the ability to capture and export data like this? Or does the Crossfeed data contain these variables?

BTW, no need for speed or bearing etc as these can be calculated from the given data.

If you can send me any data like this, then I can make it fly. ;-)

geoffmcl commented 8 years ago

@theo-armour have no problem using issues in a repo... and yes its markdown, and other options are all ok... it is for sure one of the norms I handle...

I added a new rough flightpath-r3d. It uses my modified flightpath-03.csv... so I got to understand the z,x,y coordinates... it seems the speed is based on the space between 3d points... but need to experiment more... and wish there was no pause between the land and takeoff...

Real Data set

A real Flightgear (FG) flight, flying fgfs, can be recorded. This can be done many ways, and with that, many options of what the set consists of -

A: Connecting FG to multiplayer, a fgms server instance, say mpserver01, mpserver14, ...

Now I have not yet done A:1, nor B:, but will try to get to those...

And the A:2 is a flight from LEBL to LEIG - Cessna 172P 2016-04-08 01:25:39+08 ---- 2016-04-08 02:04:10+08 LEIG 00:38:31 00:31:20 232 wpts...

That is I have not yet done the desired circuit and tracked it... but will get to that... when I prefect getting the best, safe flight path... still WIP... piloting in new hilly terrain is not easy, until you get to fully know the lay of the land...

Now this kml consists of 232 Placemarks, taken each 10 seconds. It includes a <Point><coordinates>lon,lat,alt_m</coordinates></Point>, and has an order number and timestamp <name>#5:2016-04-08 01:26:36+08</name>, etc, etc, ... loads fine in Google Earth...

Now I have my published extraction perl script, kml2xg.pl, which gives me a simple cvs/xg list of wpts - lon,lat,alt_m

But as indicated, at this time I only have the full LEBL to LEIG flight recorded, but will also to the same for the LEIG PPL circuit I really want...

And still to explore the other FG recording options. Using B: we could certainly include roll, ... -

julian date, latitude, longitude, altitude, roll, pitch, yaw, gear, flaps, ...

Will come back shortly with something... ;=))

Other Issues

Should I open individual issues for each of these, and maybe others? Or we deal here for a while...

As stated, this is looking very good... thanks...

theo-armour commented 8 years ago

@geoffmcl

>> and wish there was no pause between the land and takeoff...

your wish list = my command line

can be done in two ways:

1.uncommented line in animate to force instant jump or

  1. Edit CSV file to leave gap between start and end point.

Currently #2 is applied.

script re-titled to 3.1 and read me updated

>> Should I open individual issues for each of these, and maybe others? Or we deal here for a while...

I have added your items to the read me

Real Data Set

Can I see a sample or two of 5HZ most raw data, please?

BTW, I consider KML files to be a good lesson in obfuscation, but nevermind. ;-)

More later...

geoffmcl commented 8 years ago

@theo-armour one thing leads to another...

It seem my request for no pause, should be amended to a much more complex...

If the beginning, start at zero, accellerate, and takeoff..

However, on landing, skip the first few points, but slow down, and then accellerate to takeoff speed... join the previous... maybe quite difficult to implement...

Sort of skip an overlap of spines... rejoin at takeoff... with speed... maybe need meta data to delineate... or something... but, really thanks what now seems like a no pause... down one,,,

But that is minor... to getting a real circuit to be flown...

After many, many circuits, and circuits in the c172p...

I have now decided on a 2600 AMSL left circuit... and have flown it several times... it involves instructions like, turn left at the castle on the hill, to about 70 degrees, follow up the valleys, until over the escarpment, then pick up the A-2, do a left, and re-do the tutorial for a runway 17 approach...

So I am getting closer to generating such a flight, recorded by FG Tracker... hang in there...

KML files to be a good lesson in obfuscation

Sorry, not sure what this means... that the true data is deep in a structure... or what? Wow, like most XML implementation, you really need to know the structure to do a parse... While xml, as ascii text, offers a great way for information exchange, the problem is always knowing the details of the structure ...

No, we should accept every output possible, so long as we can read the results, and generate a form we prefer...

The simple form here, cvs,xg,... is lon,lat,alt,... ...

Sometimes, as a veteran pilot, thousands of hours PIC, or as a baby PPL with, what, at most 60 hours PIC, what qualifies a person as an instructor on such matters? Need to read the appropriate instructors handbook, for that...

For sure this will be my gut feeling of choosing the correct 2600 foot, left, circuit of LEIG is as above, and is based on all that I know, or have read to this point, but I can be real wrong...

I will have an KML file, hopefully shortly ;==))

geoffmcl commented 8 years ago

Oops, where did my reply land?

On Sat, Apr 16, 2016 at 10:16 PM, Theo Armour notifications@github.com wrote:

@geoffmcl https://github.com/geoffmcl

>> and wish there was no pause between the land and takeoff...

your wish list = my command line

can be done in two ways:

1.uncommented line in animate to force instant jump or

  1. Edit CSV file to leave gap between start and end point.

Currently #2 is applied.

script re-titled to 3.1 http://fgx.github.io/sandbox/flightpath/ and read me updated

>> Should I open individual issues for each of these, and maybe others? Or we deal here for a while...

I have added your items to the read me http://fgx.github.io/sandbox/flightpath/#readme.md

Real Data Set

Can I see a sample or two of 5HZ most raw data, please?

BTW, I consider KML files to be a good lesson in obfuscation, but

nevermind. ;-)

More later...

— You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub https://github.com/fgx/fgx.github.io/issues/1#issuecomment-210887934

geoffmcl commented 8 years ago

ok, seems the right place... why do I not see this...

theo-armour commented 8 years ago

@geoffmcl

Oops, where did my reply land?

I believe right where it was meant to land.

It seem my request for no pause, should be amended to a much more complex... If the beginning, start at zero, accelerate, and takeoff..

My guess: Once we have some 'real' data - not the data we are inventing on the fly - then most of these issues should take care of themselves

No, we should accept every output possible, so long as we can read the results, and generate a form we prefer...

Bingo!

The simple form here, cvs,xg,... is lon,lat,alt,... ...

A very good starting point. Please start with data in formats like this - for me to get going quickly. Pretty please.