Le 22/09/2011 05:03, Curtis Olson a écrit : > I have something here that I think is kind of fun. Yup, I confirm this is plenty fun :-)
It worked like a charm here. However with local weather, cold front, rough day condition, the challenge is there. Due to heavy turbs the system entered huge pitch oscillations at one mile or so in approach, but recovered stability just in time, doing a step descent (at least from chase view). Then it missed the wires, but well I think the bolter function is not there yet :-) Whit acceptable weather every thing worked as expected for the two first attempts. Then, as I was continuing this mail, the thing started to try turning the hardest possible, +/- 75deg roll and 43 AoA, ending in the water. the last attempt was aborted due to dinner time but I noticed that the switch between climbout and circle induced also a serious oscillation phase, that is AP overacting during 10 or 20 secs. (fps around 20). Then after dinner a long serial of perfect runs. I've no idea yet about what could help solving those problems. Actually there are 2 birds flying mostly the same pattern 150KTS@2000FT above the Vinson !!! That's Shrike. Some ideas: - Key 'Enter' closes the control window, it shouldn't. - The engineering HUD would be more appropriate with aircraft reference instead of f-16 like with camera reference, at least to have an idea of the controls setting. - Why not simply add the camera view with a high view number so it take place in the original numbering scheme between RIO and Pilot's view, thus keeping the usual shortcuts available ? - This kind of well formed approach would be very nice: http://www.maison-capestang.com/fg/a6/A-6E-carrier-landing-pattern-ng.png - The Downwind to Final switch is a bit rough and the gear is downed at the same time, this has a weird visual effect. - Moving target detection on mouse click and lock. - A 3D control panel in the cockpit, looking a bit custom, with red tape and big labels, like those you can see on test beds... I wonder who could do that. - Readable Nasal ??? Conclusion: AWESOME !!! Thanks a lot for sharing this, Alexis > I've been fiddling with this off and on since last fall and decided it > was time to clean it up a bit and quit hording all the fun for myself. > Basically I have taken the F-14b and created a high performance Navy > "drone" out of it. It can auto-launch from a carrier, auto fly a > route (if you've input one) and can do circle holds (compensating for > wind.) I've added a simulated gyro stabilized camera that will point > at anything you click on and then hold that view steady no matter what > the airplane does (similar to what real uav's can do.) Finally, you > can command it to return home and it will find the carrier, setup a > reasonable approach and nail the landing perfectly every time > (factoring in wind, carrier speed, etc.) > > I put together a quick web page that includes more of an explanation > and description of what the demo does. I have a link to a zip file > you need to download. This must be extracted over the top of the > existing f-14b as per the installation instructions on the following > web site: > > http://www.flightgear.org/uas-demo/ > > I'm hoping to get a few people that would like to try this and report > back on a couple things: > > - were you able to get it to work? Were there any missing files or > major blunders in the .zip file package? > > - are there places where my web page instructions stink, and can you > help me write better or more accurate instructions, especially for the Mac > > - I already know my instructions for setting up the vinson demo aren't > good, but it's been so long since I tried to do this on windows I > forget all the fgrun details. Maybe there is an easier way now? > > - finally, what do you think? general impressions? things you thought > were especially cool, or especially stupid? You probably can think of > a dozen feature requests, and I have some things in the pipeline > already. (For instance I have a refueling mode that is currently > disabled, but almost is close to working. And I've done some > preliminary work on adapting all of the auto-land logic for runway > landings.) > > - if you happen to go look at the nasal code that does all the magic, > please don't judge me (quoting Eskeletor from nacho libre) -- that was > actually a fun sub-project (for a former computer scientist.) :-) > > - Oh, and eventually I'd like to add pictures to the instructions. If > you happen to catch an especially cool looking view (weather, clouds, > time of day, sun, sun glint, scene composition, etc.) then please feel > free to send me a picture or two (or even a youtube movie) so I can > make the instructions prettier and more exciting. :-) > > If I can get this demo all cleaned up and generally running pretty > well, I have another UAS demo that is similar, but centered around the > ATI Resolution-3 airframe (which is a 92" 2.33m composite marinized > flying wing.) Then if that all goes well, I have actual embedded C > code to do much of these same sorts of things that can run on a > gumstix embedded computer (or similar.) This code is able to talk > directly to flightgear via udp packets, and has actually flown in a > couple different UAV airframes using real sensors and real actuators. > So you might see a progression developing here from pure simulation > with all the logic prototyped in nasal, to software in the loop > running pure C/C++ code, to the same software running on actual > embedded hardware (using FlightGear as reality), to the end result of > an actual real life UAV. (And I've been using drone, UAS, and UAV > pretty interchangeably here.) > > Thanks! > > Curt. > -- > Curtis Olson: > http://www.atiak.com - http://aem.umn.edu/~uav/ > <http://aem.umn.edu/%7Euav/> > http://www.flightgear.org - http://gallinazo.flightgear.org > > > ------------------------------------------------------------------------------ > All the data continuously generated in your IT infrastructure contains a > definitive record of customers, application performance, security > threats, fraudulent activity and more. Splunk takes this data and makes > sense of it. Business sense. IT sense. Common sense. > http://p.sf.net/sfu/splunk-d2dcopy1 > > > _______________________________________________ > Flightgear-devel mailing list > Flightgear-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/flightgear-devel ------------------------------------------------------------------------------ All of the data generated in your IT infrastructure is seriously valuable. Why? It contains a definitive record of application performance, security threats, fraudulent activity, and more. Splunk takes this data and makes sense of it. IT sense. And common sense. http://p.sf.net/sfu/splunk-d2dcopy2 _______________________________________________ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
