Board index Other Hangar talk

Tell us why you are happy!

Talk about (almost) anything, as long as it is no serious FlightGear talk and does not fit in the other subforums.
Forum rules
Please refrain from discussing politics.

Re: Tell us why you are happy!

Postby Johan G » Sun Jun 12, 2011 9:57 pm

I'm happy because I saw so many pilots on-line today. 73 pilots at one time! :shock: 8)

Image
Wow! 73 pilots! by Johan G, on Flickr

I'm also happy because I have registered an account on the FlightGear Wiki, mostly to be able to contribute there. :wink:
Low-level flying — It's all fun and games till someone looses an engine. (Paraphrased from a YouTube video)
Improving the Dassault Mirage F1 (Wiki, Forum, GitLab. Work in slow progress)
Some YouTube videos
Johan G
Moderator
 
Posts: 6629
Joined: Fri Aug 06, 2010 6:33 pm
Location: Sweden
Callsign: SE-JG
IRC name: Johan_G
Version: 2020.3.4
OS: Windows 10, 64 bit

Re: Tell us why you are happy!

Postby Gijs » Sun Jun 12, 2011 10:02 pm

Johan G wrote in Sun Jun 12, 2011 9:57 pm:I'm also happy because I have registered an account on the FlightGear Wiki

And that makes me happy :D
Airports: EHAM, EHLE, KSFO
Aircraft: 747-400
User avatar
Gijs
Moderator
 
Posts: 9544
Joined: Tue Jul 03, 2007 3:55 pm
Location: Delft, the Netherlands
Callsign: PH-GYS
Version: Git
OS: Windows 10

Re: Tell us why you are happy!

Postby Gijs » Thu Jun 16, 2011 3:45 pm

I just heard that I passed my examms!!!
On I go to university :D
Airports: EHAM, EHLE, KSFO
Aircraft: 747-400
User avatar
Gijs
Moderator
 
Posts: 9544
Joined: Tue Jul 03, 2007 3:55 pm
Location: Delft, the Netherlands
Callsign: PH-GYS
Version: Git
OS: Windows 10

Re: Tell us why you are happy!

Postby 439Tiger » Thu Jun 16, 2011 3:48 pm

Image
Specializing in Canadian built aircraft and aircraft flown by the RCAF

http://www.flightgearcanada.ca/ or: https://sites.google.com/site/flightgearcanada/
User avatar
439Tiger
 
Posts: 1163
Joined: Sat Jan 01, 2011 3:39 am
Location: St-Georges-de-Beauce, Québec, Canada (CYSG)
Callsign: C-CYSG
Version: 2.10.03
OS: Windoze 7

Re: Tell us why you are happy!

Postby jack » Thu Jun 16, 2011 5:54 pm

Image
For Military Aircraft, Scenery, and more, visit http://alphashangar.co.nr/

'Retired' from FlightGear as of July 2011. You can contact me via my website if you'd like to pick up any old projects.
jack
Retired
 
Posts: 1432
Joined: Wed Mar 03, 2010 12:26 am
Location: KLVK
Callsign: Alpha-J, Rescue1
Version: GIT
OS: Mac OS X

Re: Tell us why you are happy!

Postby 439Tiger » Thu Jun 16, 2011 6:46 pm

:shock: I should copyright my "Tiger Like Button" :wink:
Specializing in Canadian built aircraft and aircraft flown by the RCAF

http://www.flightgearcanada.ca/ or: https://sites.google.com/site/flightgearcanada/
User avatar
439Tiger
 
Posts: 1163
Joined: Sat Jan 01, 2011 3:39 am
Location: St-Georges-de-Beauce, Québec, Canada (CYSG)
Callsign: C-CYSG
Version: 2.10.03
OS: Windoze 7

Re: Tell us why you are happy!

Postby curt » Tue Jun 21, 2011 2:33 pm

Image

Image

Image

Great first flights of our UAS airframe (the first one out of the molds). The wings and vstabs are unpainted. Hope to have a FlightGear version of this released soon (with full autonomous flight capability). Here are some more pictures:

https://picasaweb.google.com/1055786701 ... feat=email
curt
Administrator
 
Posts: 1168
Joined: Thu Jan 01, 1970 1:00 am
Location: Minneapolis, MN

Re: Tell us why you are happy!

Postby redneck » Tue Jun 21, 2011 3:09 pm

I'm very interested in full autonomous flight. That's mainly what lead me to develop FGFSFMC, which I'll be renaming in the next release, as it has been brought to my attention that I have made it too smart to model just an FMC. Hmm. Guess I got more incentive to work on what I think will be the final release now. I'd like to compare my program to whatever you develop for the fully autonomous flight. Or, if you're experienced with Java, and would like to take a look at my code, or even use some of it, you're welcome to. I'll just warn you, I don't annotate my code much, if at all, but I like to comment out stuff that doesn't work, instead of deleting it. Furthermore, the flow of the code does not match the phases of flight as they are encountered chronologically, as I was unable to get it to work right when I made it that way, and have pretty much scrambled the order of conditions pretty well. If you need help finding something, just let me know. I'll be releasing it under the GPL in the next release (past releases were made free under no license at all). Not that GPL makes my software any more secure than no license at all, AFAIK, but just to keep it in the same open-source spirit as FG.

Being that you are the one in charge of the FG project, I know you have much more coding experience than I do, so I'm sure you could make something to fly this plane on its own much better than I could. Furthermore, I expect you'll probably do this in Nasal, instead of Java, so you can build it directly into the aircraft, instead of making it an addon, so it can run faster.

Finally, I would also like a look at your code, once you've completed it. Since I have a little coding experience, and Nasal seems fundamentally the same as Java, just with a different syntax, I just might be able to understand it. I'd be very interested to see what approach you'll take. Something tells me it won't be anything like mine. The automation will control the flight surfaces directly, instead of telling the AP what to do as I have, right?

Oh, and I got another question for you. How much will the automation be able to do? I expect it will do more than simply follow the Route Manager, as the AP we have does that just fine. Will it calculate top of descent, and begin descent at that point, adjusting descent rate to maintain a 3-degree descent profile? What about autoland? Auto takeoff? Anything I may have missed?
Call Signs: redneck, ATCredn (unspecified freq atc)
FGFSCopilot
FGFSCopilotATCEdition
System Specs
Model: Alienware M15x, OS: Windows 7 Professional 64-bit, RAM: 3 GB, CPU: Intel i3 quad core at 2.4 GHz, GPU: Nvidea GeForce GTX 460M 1.5 GB GDDR5
redneck
 
Posts: 3617
Joined: Mon Feb 02, 2009 3:17 am
Location: Pennsylvania, USA
Version: 240

Re: Tell us why you are happy!

Postby redneck » Tue Jun 21, 2011 3:51 pm

It's related b/c he's happy about the great first flights. I don't think I've derailed the topic too much. Apologies if I have.

Guess I'll add in a more related subject. I will be happy when I learn how to properly release my software under the GNU GPL, and can release the latest version of FGFSCopilot. Yes, it's finally being renamed. Which reminds me. I think I forgot to change the titles of the windows. Should be a quick fix. Does that count as on-topic? Not happy now, but I know I will be, hopefully soon.
Call Signs: redneck, ATCredn (unspecified freq atc)
FGFSCopilot
FGFSCopilotATCEdition
System Specs
Model: Alienware M15x, OS: Windows 7 Professional 64-bit, RAM: 3 GB, CPU: Intel i3 quad core at 2.4 GHz, GPU: Nvidea GeForce GTX 460M 1.5 GB GDDR5
redneck
 
Posts: 3617
Joined: Mon Feb 02, 2009 3:17 am
Location: Pennsylvania, USA
Version: 240

Re: Tell us why you are happy!

Postby curt » Tue Jun 21, 2011 4:46 pm

I have been fiddling with auto take off, and auto landing. Getting the approach to be smart and optimal is hard. I have something that works pretty well for light winds, but have been trying to make it smarter for higher wind scenarios ... so right now everything is torn apart and not working so well. I've also done an implementation of constant radius circle holds. Anyway, I just need to find some time to get back at the approach code and get that working and then I plan to make the airplane available. (And I know the "release early, release often" thing, but this is part of a commercial project so I have some different constraints to consider on this one.)

Redneck: on your FMC, often it is not too hard to make something way smarter than what you would find in a typical cockpit. :-) Then you have to decide if you want realism or all the capabilities ... in this sense the UAS stuff is fun because you do have constraints of what you can do with real world sensors and embedded compute power, but you aren't limited to mimicking only what someone else has done before.

On Java: I've never touched it. I'm sure the basic syntax would be pretty familiar, but I get lost in all the API's and supporting libs ... knowing what's available and what you can do I think is where most of the mental energy would go in learning it.
curt
Administrator
 
Posts: 1168
Joined: Thu Jan 01, 1970 1:00 am
Location: Minneapolis, MN

Re: Tell us why you are happy!

Postby redneck » Tue Jun 21, 2011 5:42 pm

curt wrote in Tue Jun 21, 2011 4:46 pm:Redneck: on your FMC, often it is not too hard to make something way smarter than what you would find in a typical cockpit. :-)

Interesting. I guess people in RL haven't put as much faith in flight automation as I have for this sim. The autolander I made isn't quite as intelligent as I'd like it to be, but it gets the job done. Call it laziness if you want. I just like watching a computer do something that would normally require a human being.

curt wrote in Tue Jun 21, 2011 4:46 pm:Getting the approach to be smart and optimal is hard. I have something that works pretty well for light winds, but have been trying to make it smarter for higher wind scenarios

My program isn't too smart when it encounters high winds either. Since my program doesn't directly control the flight controls, but tells the AP how to fly the plane, I don't think I currently have the skill to implement some method for approach in high winds. So, what I did was add in a condition to determine if autolanding will be attempted. Basically, if the crosswind component is at least 15 kts, then it's a no-go. So, it will deactivate the AP as well as itself, and let you know that it has done so to not only give control back to the pilot, but make the pilot know that he must land the airplane.

With your work being a commercial project, Idk if you have the option of simply avoiding the high wind scenarios like I have. Having a program that can land a plane by itself in high winds would be nice. I just don't have the understanding of exactly how I manage to make a high crosswind approach and a safe landing manually. Basically, I know I turn into the wind so that I can fly on a straight course to the runway. Exactly how much to turn is usually based on visual cues, and is a constant adjustment for me. Then comes landing. Decrab too early, and you get pushed away from the runway. Decrab too late and... well you won't get a chance to decrab before you touchdown. Basically, when the runway appears to be just a certain distance below me, I kick the rudder in whatever direction I need to straighten out, applying a little more aileron to keep the wings level, and hold it steady until I touchdown. If I could convert all that to the necessary numbers and variables, I'd at least be able to write up some pseudo-code just for this.

But then comes the issue with efficiency. I have found that it takes a couple seconds or so for my program to run a check through all the conditions and execute code, or not execute code. I believe there are two ways to deal with this. One method is to write the code more efficiently, something I'm not exactly sure I know how to do. Then, there's heavy flight testing. For example, my autolander code was waiting too long before flaring the plane. So, I wrote the code so that, if the plane is about (I think) 75 ft or so higher than where I actually want it, then it will execute the code. So, once I reach that altitude and satisfy the condition, I have to wait for the program to actually check that, and then by the time it gets there, I will have lost the extra 75 ft or so, putting my plane exactly where I want it for it to start the flare. This method requires LOTS of flight testing however, and that can get tiring sometimes, and VERY annoying when it fails. I'm not sure if you even want to consider this method, but I thought I'd let you know it's out there.
Call Signs: redneck, ATCredn (unspecified freq atc)
FGFSCopilot
FGFSCopilotATCEdition
System Specs
Model: Alienware M15x, OS: Windows 7 Professional 64-bit, RAM: 3 GB, CPU: Intel i3 quad core at 2.4 GHz, GPU: Nvidea GeForce GTX 460M 1.5 GB GDDR5
redneck
 
Posts: 3617
Joined: Mon Feb 02, 2009 3:17 am
Location: Pennsylvania, USA
Version: 240

Re: Tell us why you are happy!

Postby curt » Wed Jun 22, 2011 3:15 am

I've done a fair bit of work computing the ideal heading in order to fly the desired ground track in wind. What I would like to do is adjust the glideslope based on the wind vector. In a small RC scale airplane if your approach speed is 25 kts and you are flying into a 22 kt head wind, and you fly a standard approach with a 3 degree glide slope, you fly way out and then inch your way back in, perhaps at a medium cruise throttle setting. You are basically a kite without a string.

What I would like to do is adjust the whole approach decent all the way back up through the down-wind leg to compensate and fly a much steeper approach if there is a significant headwind component. This isn't something you'd typically do in full scale aviation, but it is something you would do as an RC pilot and it make sense in the context of this project. This is especially important if you are auto-landing a glider that has flown a full mission and the batteries are expired -- you have to be really careful with energy management or you come up way short or overshoot, neither of which are good options.

Oh, I'm also happy because I was playing around with some hardware PWM signal generators on our embedded flight computer and was able to find the right configurations to drive all 4 outputs with the 13Mhz clock. This gives me about a 0.1us granualarity across a 1000 - 2000us pulse range @ 50hz. That means I can specify about 10,000 unique positions across the full range of servo motion ... and the PWM signal is generated in hardware so it is rock solid exact. I don't know if a digital servo can respond to that small of a step size, but it's nice to know that we can generate the signal that well anyway. To put it in perspective, if we were driving a pan/tilt camera through 180 degrees of rotation, we would have better than 0.02 degree accuracy (assuming the servo itself could resolve the input signal with that level of precision.)
curt
Administrator
 
Posts: 1168
Joined: Thu Jan 01, 1970 1:00 am
Location: Minneapolis, MN

Re: Tell us why you are happy!

Postby Gijs » Sun Jul 03, 2011 8:16 pm

Gijs wrote in Thu Jun 16, 2011 3:45 pm:I just heard that I passed my examms!!!

And last Friday I actually got the proof: my certificate! At the graduation ceremony I had to come up twice. First to get my certificate, then to accept the prize I won, by writing the best profile paper of my school! Oh, and the same day my mum's birthday was celebrated :D
Airports: EHAM, EHLE, KSFO
Aircraft: 747-400
User avatar
Gijs
Moderator
 
Posts: 9544
Joined: Tue Jul 03, 2007 3:55 pm
Location: Delft, the Netherlands
Callsign: PH-GYS
Version: Git
OS: Windows 10

Re: Tell us why you are happy!

Postby polly » Sun Jul 03, 2011 9:07 pm

Congrats on all, good fortune to you.
User avatar
polly
 
Posts: 956
Joined: Thu Nov 04, 2010 3:45 pm

Re: Tell us why you are happy!

Postby 439Tiger » Sat Jul 09, 2011 4:07 am

I'm happy because I will be starting a new job in a week's time :D

However that means my livery making will be slowed down a bit :mrgreen:

I guess I should do a blitz next week... :shock:
Specializing in Canadian built aircraft and aircraft flown by the RCAF

http://www.flightgearcanada.ca/ or: https://sites.google.com/site/flightgearcanada/
User avatar
439Tiger
 
Posts: 1163
Joined: Sat Jan 01, 2011 3:39 am
Location: St-Georges-de-Beauce, Québec, Canada (CYSG)
Callsign: C-CYSG
Version: 2.10.03
OS: Windoze 7

PreviousNext

Return to Hangar talk

Who is online

Users browsing this forum: No registered users and 6 guests

cron