Page 3 of 4

Re: Firmware and Params for autonomous flight (Link to firmware added)

Posted: Sun May 21, 2017 12:43 am
by freaad
EK3_ALT_SOURCE: Primary altitude sensor source
I use: EK3_ALT_SOURCE = 2
EK3_GPS_TYPE: GPS mode control
Note: This parameter is for advanced users
This controls use of GPS measurements : 0 = use 3D velocity & 2D position, 1 = use 2D velocity and 2D position, 2 = use 2D position, 3 = Inhibit GPS use - this can be useful when flying with an optical flow sensor in an environment where GPS quality is poor and subject to large multipath errors.
0 GPS 3D Vel and 2D Pos
1 GPS 2D vel and 2D pos
2 GPS 2D pos
3 No GPS
I use: EK3_GPS_TYPE = 1
(Don't know why, I certainly have tried setting 0, but probably with bad results)
I am wondering what bad result did you face it? I confuse this part. You are using the 2d vel and 2d pos but altitude source is GPS.
My short knowledge, it is hard to understand about this parameter source. I will try to 2D vel 2d pos.
Also did you try to use RTL mode? When I use the RTL mode it goes up even if I checked that the altitude parameter gives 0(I am relative distance setting ).

Re: Firmware and Params for autonomous flight (Link to firmware added)

Posted: Sun May 21, 2017 1:00 am
by Hendrik
Rtl doesn't work for us either.

Also LAND instruction in a mission (waypoints) doesn't work.
We have only waypoints (and servo positions). Landing is done by going to 0m06. We than manually switch off with emergency stop (channel 8 in my params)

Re: Firmware and Params for autonomous flight (Link to firmware added)

Posted: Sun May 21, 2017 1:08 am
by Hendrik
About EK3_GPS_TYPE:

Setting 0 = 3d velocity and 2d position. (Not 3d position!)
We don't want a calculated altitude but be as close as possible to gps measured height.

Re: Firmware and Params for autonomous flight (Link to firmware added)

Posted: Sun May 21, 2017 3:52 am
by freaad
Land mission worked for me. Loitering also worked.
Rtl seems dangerous both .
I have big accident middle of rtl. I need to make another UAV again.
Thanks. After test normal mission I will contact you. Next week is my final exam.

Re: Firmware and Params for autonomous flight (Link to firmware added)

Posted: Mon May 22, 2017 11:11 am
by amolad
Interesting..RTL didn't work for us either. We were thinking that the parameters were messed up! Okay gonna give it another try in Auto and Guided mode.
Were you guys successful in Guided?

Re: Firmware and Params for autonomous flight (Link to firmware added)

Posted: Mon May 22, 2017 6:58 pm
by Hendrik
We fly in auto mode.

Re: Firmware and Params for autonomous flight (Link to firmware added)

Posted: Tue May 23, 2017 11:39 am
by Debupt
hiļ¼ŒHendrik~

Recently I use your Firmware and Params for testing. When I test the Loiter mode, the drone just cannot and still drifts for one direction. I find that we use the different frame type( quad and hexa) , I wonder whether this is the problem? ( I change the frame type in my Mission Planner)

And could you please give some advice for Loiter mode.

I just send RMC,GGA and VTG to pixhawk every 150ms.

The control process is taking off with Stabilize mode, changing to Loiter mode for fixing position.

Re: Firmware and Params for autonomous flight (Link to firmware added)

Posted: Tue May 23, 2017 2:20 pm
by Hendrik
@debubt

What does the plot on the map in mission planner looks like?

If it's displaying the same distortion, then the pixhawk is not correcting enough. Try increase P value for position.

if it's displaying a different path then you see it has to be in the positioning itself.

Please response with screendump.

Re: Firmware and Params for autonomous flight (Link to firmware added)

Posted: Thu May 25, 2017 12:06 pm
by Debupt
@Hendrik
screenshot.jpg
screenshot.jpg (144.54 KiB) Viewed 11589 times
When I tried to set the coordinates on QT software that uses these coordinates and converts it into nmea protocol, the drone in the Mission Planner will go to the coordinates set by me. However, the drone will oscillate about the coordinate point (up and down) and after a few milliseconds of delay, it will then go to the exact coordinate point.

Do you know the reasons behind this delay? Is it caused by the parameters?

Re: Firmware and Params for autonomous flight (Link to firmware added)

Posted: Sat Jun 17, 2017 6:43 pm
by Debupt
@Hendrik, @freaad, @amolad

The test video is: https://youtu.be/cxoyxIBi7ac

The problem is that it still drifts.

Also the Yaw value drifts seriously from 0 to 360 and sometime the drone just goes up fiercely even the altitude value is normal.

Hope for some advice~~