Re: [Paparazzi-devel] InsideGreen() or non-circular mission boundary?

[Felix Ruess]

Hi Simon,

thanks for testing. Merged the fix into v5.0 and tagged it as v5.0.4_stable.

Would be great if you could test master as well (lots and lots of updates and fixes compared to v5.0). Would be good to know what issues you encounter when switching to master, so we can improve the error messages where it makes sense and fill in the guide on how to upgrade: https://wiki.paparazziuav.org/wiki/Release_Upgrades

Cheers, Felix

On Wed, Mar 5, 2014 at 12:37 PM, Simon Liebold <address@hidden> wrote:

All,

OK, I initially tested with master but I got too many scary looking compile errors. My configuration probably doesn't match it.

With v5.0_fix_inside it works perfectly. Just as one would expect. If everything works as planned, I will test it in the air over the weekend. And - when feeling brave enough - I will eventually switch to master.

Thanks all. I learned a lot about the Paparazzi internals in the past few days.

Simon

Am 05.03.2014 07:57, schrieb Simon Liebold:

Hi Felix,

what paparazzi_version are you using?

Yes, the latest stable 5.0 branch.

Could you try with latest master?

Sure, glad to do that. I will try that in the course of the day.

Simon

Am 04.03.2014 23:38, schrieb Felix Ruess:

Yes, there are still some open issues, but v5.2 is planned soon... will probably create a v5.1_testing branch after some of the outstanding pullrequests are merged.

So testing latest master is welcome!!!

On Tue, Mar 4, 2014 at 11:36 PM, Christophe De Wagter <address@hidden> wrote:

Sounds like time to consider a new release like 5.2?

On Mar 4, 2014 11:29 PM, "Felix Ruess" <address@hidden> wrote:

Simon,

what paparazzi_version are you using? Could you try with latest master?

If you are using v5.0 stable it might be issue 602, which should be fixed in master.

I also created a new v5.0_fix_inside branch, with that fix backported. If you are on v5.0 stable, could you please test if that resolves your issue, then I will merge it in v5.0.

Cheers, Felix

On Tue, Mar 4, 2014 at 11:14 PM, Gautier Hattenberger <address@hidden> wrote:

Hi,

I don't have much time right now, but your flight plan actually works as expected for me ...

Gautier

Le 04/03/2014 22:38, Simon Liebold a écrit :

Hello,

So, I changed the order of the polygon corners (now clockwise) and simplified the flight plan a bit. Still, it did not change the outcome.

What puzzles me is that it keeps jumping to the "Center" Block(8) directly after I started the simulator (see attached screenshot) while the simulated airframe is still stationary on HOME. Again, manually switching to any other block will result it jumping back to Block(8) a split-second later.

Now I looked at the generated code (flight_plan.h) trying to understand what is happening with the inputs (x=0.4375; y=0.5):

static inline bool_t InsideGreen(float _x, float _y) { \
  if (_y <= 312.6) {
    if (_y <= -179.7) {
      if (_y <= -569.2) {
        return FALSE;
      } else {
        float dy = _y - -179.7;
        return (-845.1+dy*-3.047754<= _x && _x <= 474.0+dy*0.338852);
      }
    } else {
      float dy = _y - 312.6;
      return (-593.3+dy*0.511548<= _x && _x <= 640.8+dy*0.338852); // It should return this line.
    }
  } else {
    if (_y <= 608.3) {
      float dy = _y - 608.3;
      return (-442.0+dy*0.511548<= _x && _x <= -442.0+dy*-3.661819);
    } else {
      return FALSE;
    }
  }
}

And I don't see why it would not return the line in the middle but jumps to one of the "return FALSE". I have to say, I am not a C expert (yet)...


Simon

Am 04.03.2014 12:26, schrieb Simon Liebold:

Hi Christophe,

Tip: make sure to define your polygons clockwise.

You were right. I had defined it counter-clockwise. But I just briefly tested it is still giving the same results. I will investigate this later when I have more time.

Thanks so far for the hints.

Simon

On Tue, Mar 4, 2014 at 9:04 AM, Simon Liebold <address@hidden> wrote:

Hi Gautier,

thanks for offering help. I did not know that it auto-generates these functions. Interesting.

I tried to use the exception with and without the hight limitation. Still it will jump to "Center" immediately after starting up. If, during simulation, I manually switch to any other block it will jump right back to "Center". It looks as if

!(InsideGreen(GetPosX(),GetPosY()))

is TRUE all the time (not InsideGreen).

I am attaching my flight plan. Maybe you can find my mistake.

Simon

_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

--
Simon Liebold | address@hidden | Key ID: 0xB59D1EC6 | CAcert Assurer
Cell +49 151 550 671 53 | Fon +49 69 173 265 300 | sip:address@hidden

_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel