@Ilya_Smolnikov wrote:
The:
holibro kakute f7 aio
quad 250
- Weight 1000 grams (2 lbs)
The problem is vibration, ECF error, GPS failure, dispersion error, but according to the logs, the vibrations are small.
I just can’t handle this problem. What I just didn’t do …
I decided to solder the gyroscope, remove the ICM 20689 and solder the mpu6000. but it didn’t work, tell me why ???
LOG
Posts: 6
Participants: 2
@Berto wrote:
Hello Everyone,
I’ll try to keep this short-ish; I’m a new user, and hopefully not the first to be a little overwhelmed by the scale of choice available. There’s an awful lot to read and I’m getting through it, but at this point I need a little nudge in the right direction, please…
The project is for a water based craft that will fly using a live hydrofoil wing requiring stability input for pitch and roll of the foil and height of the hull above water. The craft will also need to loiter/hold position in moving water before execution to follow a path.
I ‘think’ the most applicable firmware/vehicle type to start with will be a quadplane, substituting vertical rotors for thrusters to loiter/hold position and then transition to forward ‘flight’ using hydrofoil elevons and propeller.
I’m banking on the ability to customise the function of each input/output, but as there doesn’t seem to be a vehicle type already listed I need to to start with something.
Your thoughts and opinions are much appreciated, as is the wealth of information in the forum!
Kind regards,
Berto
Posts: 1
Participants: 1
@TheTroll wrote:
Hello
When I’m connecting usb cable from drone to QGC based on Android device it connects automatically but only with 56k uart speedIs it any possibilities to configure this to 9600?
Because I’m using low speed telemetry link.
Thanks!
Posts: 1
Participants: 1
@tksystems wrote:
What GNNS formats does Ardupilot support?
I am using two ZED-F9P that output UBX-NAV-RELPOSNED (pg 158 of Interface Manual)
I was wanting to use the heading information and disable all magnetometers.
Thank you!
Posts: 1
Participants: 1
@mike_d wrote:
Recently I stumbled upon a feature in another flight control system called FlyTogether by MyFlyDream. Essentially, it paints an indicator in your OSD to show another aircraft’s heading, relative altitude, and relative distance to your aircraft, and which way you need to fly to find them. Video in this link.
Technically, it seems that MFD did this by using 2 layers of OSDs. Both parties needed to be using MFD antenna trackers, and you would physically connect up to 8 of these antenna trackers together. The antenna trackers would communicate their aircraft’s coordinates, and since your video went through the antenna tracker, the tracker would paint another OSD onto your video with the aircraft locations.
Since we’re in the age of ChibiOS, flight controllers with built in MAX7456 chips, and async mesh on rfd, is this something that’s physically possible, especially without using antenna tracking hardware? I know the OSD chips are character based, so it would probably be difficult to get the same fidelity.
Posts: 1
Participants: 1
@wicked1 wrote:
I have a copter I really like… Little 3" prop microcopter. 1402, 8100kv motors. 2s. 190grams.
I built a twin of it, only it’s lighter on a different frame. Also 3". same props and motors. 160 grams.
With the heavier copter I can fly for 7 minutes w/ battery to spare.
With the new lighter copter I can only fly for about 5.5 minutes.I have two theories. Either this new ESC is much less efficient. Both blheli_32, bot 30 amp. Both have the same settings. Old one is Aikon AK32. New one is Lumenier Micro Razor.
Or, the frame geometry. The old one is kind of H frame, while the new one is X. They both have braces going from the front motor to the rear. I’m no expert in fluid dynamics or anything, but the way I’m imagining things, there would be turbulent air around the arms. The area between the arm and the front to rear brace is pretty tight on the X framed copter… Maybe that entire area is nothing but turbulence and not much air is going down, while the H frame has more space between the arm and the braces. I’ll post a pic of what I’m thinking.
My other reason for this theory is the top speed seems to be lower. But, that is not a scientific observation… Just a few full throttle runs.
Here’s what I’m thinking… Turbulent air in red, blue is the downward air.
I wouldn’t expect either thing to make about 20% difference!!
I do have another Aikon ESC, but really don’t want to rebuild the copter… (ESC is of course at the bottom of the stack… Have to completely disassemble and rebuild to replace it).
Anyone else think my frame/airflow theory might be correct?
Posts: 1
Participants: 1
@Pogo wrote:
I’m fairly new to threads, so I have a query. I notice that in AP_Terrain,
void schedule_disk_io(void)(and other related functions) runs in one thread, andvoid io_timer(void)(and other related functions) runs in another thread. Both threads access thevolatile enum DiskIoState disk_io_statevariable to coordinate disk access.Is access to the
disk_io_stateenum an atomic operation? I imagine it must be to prevent race conditions.Thanks.
Posts: 1
Participants: 1
@Raz_St wrote:
Hello,
I’m controlling my pix4 using RaspberryPI and python code (DroneKit). I’ve been using the code for a while with no major issues. My code basically switch into guided_noGPS mode then set throttle to 0.6 (takes off) until it reach 1.5 meters than I set the throttle to 0.5, change the roll a little bit than land.
Yesterday, I run this script and once it connected to the PIX the drone got full power into the engines (high throttle) and flew to the sky until I got manual control and landed it.
I tried a few more times and it went back to normal behavior but then once again, after connection to the PIX the drone got very high throttle, it hit something and crashed.
Any idea what can cause the throttle to (sometimes) be so high even if I sent it 0.6?
Not sure if related, but A few days ago I had an electrical short in the drone when I accidently connected the “+” and the “–” of the ESC power connection split board. but it looked like the drone was flying fine manually (in stabilize mode).
log file can be found here.
Setup:
pixhawk4
ArduCopter 3.6.11
Mode: Guided_NoGPS
ESC’s are clibratedThanks,
Raz
Posts: 1
Participants: 1
@korakora wrote:
i have very strange experience with WP Mission which i never ever had before
but first let me start to describe with this points
- i don’t have any log because i’m using omnibus with “logging disable”
- i’m sure i’m not making any mistake (i attach my prove)
- no gps glitch (all perfect) also i’m attach the picture
- because i don’t have any log, so i just have is realtime flight capture
as point number 1, then yes i realise i’m asking a guessing theory
but it’s fine better than no clue at allso, this is my questions is
this is the real WP i was create it (i’m upload directly from the FC it self)
nothing wrong isn’t it?
but this what happen in the real flight as we can see it should turn once once
but it make turn “forever” it was out from that circle because i take a control manual (out from AUTO)
*fastforward 16x
please any “even” guessing theory are welcome, i’m afraid this happen again (lucky this time still in my range to take over)
1/5/2020 18:53:13 : omnibusf4pro 00560037 3548500C 0373753
1/5/2020 18:53:13 : ChibiOS: 1d9d1159
1/5/2020 18:53:13 : ArduPlane V4.1.0dev (6ff303ad)MP Version 1.3.70
Posts: 1
Participants: 1
@chris661 wrote:
have 2 Beitian bn-880 each one with Omnibus F4 and both have similar problem: need a lot of time to get fix. Before I installed Copter 4.0 I get enough sattelites after around 30 seconds. I think it’s related to the GPS, because I do not have this problem with Radiolink TS100 Mini M8n on a Pixracer with Copter 4.0. It’s like this GPS is not able to save the last used GPS position. Anyone else with this problem? To understand better, is last used GPS position stored on GPS or and on Ardupilot? Any parameters to check?
Posts: 5
Participants: 3
@twistedwings wrote:
configured parameters under “Follow Me”
switched to “Parameters”
switched back to “Follow Me”
ArduCopter 4.0.0
QGC Windows daily
Posts: 1
Participants: 1
@TheTroll wrote:
Hello!
I’d just trying to build master branch of ardupilot.
After commands
Waf configure - - board=omnibusf4pro
Waf planeI have error message (on linkage stage I suppose) saying
region flash overflowed by 188 bytesAt the same time it compiles successfully for matek f405 wing board.
Could you please explain how to improve this
Thanks
Posts: 1
Participants: 1
@twistedwings wrote:
After switching from loiter to acro the copter drops.
I assume I have to modify ACRO_THR_MID which is set to 0.
What does the range of 0 to 1 mean ?
ArduCopter 4.0.0
Posts: 5
Participants: 2
@sridharanNH wrote:
I have been fighting the joystick problem for awhile and I was wondering if someone can help. My setup is this, on a Windows 10 PC: MP(3.1.70)<->AP/SITL(4.0.0 stable)<->X-Plane 11.41. I have made sure that X-Plane does not expect any direct joystick input. I start AP/SITL (–model xplane), then start X-Plane, and finally start MissionPlanner with the following arguments: /joy “T.Flight Hotas X” /type plane. After connecting MP to AP/SITL, I used the Flight Data|Actions|Joystick screen to setup the Roll, Pitch, Yaw and Throttle channels and enabled it. There were no error messages or diagnostics. When I manipulate the joystick, MP does not send any MAVLINK_MSG_ID_MANUAL_CONTROL messages to AP/SITL - I have verified this by setting a breakpoint in AP/SITL where it handles that message. Do I have to setup some MP config param or something? I see a .rlog file generated, but I don’t know how to decode it. Thanks a lot in advance for any help.
Posts: 1
Participants: 1
@twistedwings wrote:
Follow mode is set to relatvie altitude of 10m above:
!
Logged altitude is this:
Is this unprecise altitude sent from the GCS or a firmware problem ?
ArduCopter 4.0.0
QGC Android daily 2020-01-03
Posts: 2
Participants: 2
@jwcalvert wrote:
I decided to reload all my software including Eclipse, etc,etc,etc.
And Ardupilot 4.0.0 code for the rover.
But now Eclipse is ridiculously slow to build the first target.
6 hrs to get to 19%.
Any thoughts/help please.
Posts: 2
Participants: 2
@jwcalvert wrote:
18:27:23 **** Build of configuration Default for project ardupilot ****
“c:\cygwin64\bin\python3.8” waf clean
The project was not configured: run “waf configure” first!
“c:/cygwin64/bin/python3.8 waf clean” terminated with exit code 1. Build might be incomplete.18:27:25 Build Finished. 0 errors, 0 warnings. (took 1s.926ms)
I just got the above error message and have NO idea how to run “waf configure”.
Please help.Prior to re-installing Eclipse my rover builds were easy with no issues.
Now, they are a nightmare and none ever complete.Thanks
Posts: 1
Participants: 1
@vosair wrote:
Can anyone confirm that this OSD works with 4.0 and cube black? I have it connected for the to serial 2 and set it to Mavlink 1. I get video output where it shows the battery as 0v, on time, and a “No Data” message flashing. Do I need the FTDI adaptor to get it to configure properly? Any advice is much appreciated.
I followed the wiki page for connection https://ardupilot.org/copter/docs/common-minim-osd-quick-installation-guide.html
This is the exact model of OSD I’m using https://www.readymaderc.com/products/details/rmrc-micro-minim-osd
Posts: 1
Participants: 1
@crgchyn wrote:
been working on this for a while. running ver 3.6.12 this copter flys great for about 9 min then becomes unstable and just crashes. im running a 5400 mah battery. the battery and current output look ok in the logs. Does anyone with a bit more experience that may see something I am missing in the logs they are attached. I was running a auto tune this flight had to do a rtl as it got a little far away then became unstable about 30 ft from landing and crashed pixhawk logs
Posts: 6
Participants: 3
@Hacky wrote:
I am using ArduCopter on a Kakute F7 Mini since July 2019 (at that time AC 3.6.9 was current stable), meanwhile it was updated to 3.6.10 and 3.6.11. Flying without problems. Starting from 3.6.12 there was a seperate build especially for the “Mini” but I skipped 3.6.12 and updated to 4.0.0 (still on the build for Kakute F7 - non-Mini). I guess, the only difference is the default setting for the logging as the “Mini” has no microSD card slot. My setting so far was LOG_BACKEND_TYPE 0 (no logging).
Since I updated to AC 4.0.0, sporadically the flight controller reboots and after the reboot I see the message “PreArm: Internal errors (0x800)” and “watchdog_reset”. Furtunately, so far this was only before or after flights and not in-flight. I am aware that this would end with a crash when it happens inflight.
As I had to get through some flight tests today, I (thought I) disabled the watchdog with BRD_OPTIONS 0 - aware that this is a bad workaround. After the (successful) flight and disarming, again the FC rebooted and I saw the same message again.
So, although BRD_OPTIONS was set to 0, the watchdog still seemes to trigger a reboot.By searching for the root cause, I played around with the logging setting. When setting the LOG_BACKEND_TYPE to 1 (file) it reports “Bad logging” as expected and the watchdog reboot also happened while the copter was armed and motors running (copter fixed on the ground) - causing them to stop .
After I learned, that there is a “block logging” option (LOG_BACKEND_TYPE 4) that should work also on Kakute F7 Mini (writing to integrated flash memory), I switched to this logging type. Until two more tests, the watchdog reboot was not yet seen but I am not yet confident that this will fix the watchdog issue.
Now the problem is, that even with block logging, I still get no logfiles. When I try to retrieve a log, mission planner only reports:
Getting list of log files... Error:System.TimeoutException: Timeout on read - GetLogEntry at MissionPlanner.MAVLinkInterface.GetLogEntry(UInt16 startno, UInt16 endno) at MissionPlanner.MAVLinkInterface.GetLogList() at MissionPlanner.Log.LogDownloadMavLink.<LoadLogList>b__12_0()The AC 4.0.0 release notes mention that there was also some kind of rework on the watchdog, so there must be something that triggers it newly on Kakute F7 and I would like to find it out…
I did not yet switch to the dedicated Kakute F7 Mini build as this would mean, I would have to redo configuration from scratch. But if you can give me good reasons why this makes sense (are there other differences than only the logging defaults?), I will go that way…
Posts: 2
Participants: 2
!
