@dannyg wrote:
Hi everyone,
Right now I am try to control a Pixhawk UAV using a ground computer sending setpoint commands to APM. However, right now my telemetry radio setup is inconsistent. It continuously drops connection making this offboard command impossible. I am currently using mRo SiK Telemetry Radios for communication.
Is there a better out of the box solution for a fast/reliable connection for offboard control?
Thanks!
Posts: 3
Participants: 3
@OlivierB wrote:
Congratulations to talented Ardupilot community member Andrea Belloni (aka anbello on Github), for winning Ardupilot’s “Best Contribution of the Month” award for March 2019!
Andrea was nominated and won the prize for his work on indoor autonomous flight with Arducopter, ROS and Aruco Boards Detection. This is the first project (with wiki documentation) that demonstrated a fully functional Ardupilot system taking advantage of the newly developed interface for vision position estimation in a gps denied environment.
For those that may not be familiar, Aruco boards are boards with patterns that can easily be printed and used to infer, via vision algorithms, the 3D pose of an aerial vehicle with a camera and companion computer.
Building upon and integrating these algorithms with OpenCV, ROS, Mavros, Andrea successfully demonstrated autonomous vision based flight with Arducopter, on a small copter prototype equipped with a Pi3 companion computer and pi camera module he designed and built.
Andrea’s work and prolific interactions with others interested in this particular field (e.g. dev team member ppoirier, former contribution of the month winner chobitsfan, Subodh_mishra and many others) also provided a communication platform and foundation for others to further experiment and build upon, given his code contributions, documentation, and easy to build prototype. See for instance dev team member ppoirier related work on Monocular Visual Inertial Odometry with an inexpensive skyviper, or Andrea’s earlier work on the subject.
As one developer has said and to paraphrase, “This young developer-experimenter represents all that the ArduPilot community is about”, and the dev team is delighted to recognize his pioneering work.
The prize for the month was $200 for the top contributor. Many thanks to those who donated to ArduPilot, including our Corporate Partners, totalling 57 to date! If you are a company and wish to donate a prize for an upcoming month, please email the partners email list (partners@ardupilot.org).
Congrats again Andrea!
Posts: 6
Participants: 5
@rolyexpress wrote:
Hi guys,
I’d like to know how to tell if my voltage sag is within normal parameters.
When my quadcopter is at 50% battery (4S, according to my FlightDeck script on Taranis) I’m getting 14.5V. This is while in auto mission & moving forward.
I have a 680mm Quad frame with 2kg AUW. It’s running on 4S 5200mAh (35C/70C Burst) from Lumenier. I’m swinging 15"x5.5 props and four T-Motor MN3508-380kv. I’m using Mauch Power modules.
Off the bat: I know my quad is heavy for the thrust on 4S. Also my batteries are slightly puffy, I’ve never had batteries puff so I can’t discern the gravity of the swelling.
Thanks in advance!
log file: https://www.dropbox.com/s/7v5kebof0g2bq1o/2019-05-09%2015-37-02.bin?dl=0
Posts: 2
Participants: 2
@ivanh wrote:
I have replicated the following behavior with MP 1.3.62 and 1.3.66…
When installing firmware (using ArduCopter v3.6.8) I reply “No” to the “Upload ChibiOS” prompt because I want Nuttx installed. However, even though I reply No to this question it is installing ChibiOS instead of Nuttx. The prompt works correctly on previous firmware version installs, allowing me to flash with Nuttx. However AC 3.6.8 firmware refuses to take No for an answer.Has Nuttx been completely eliminated? The fact that there is still a prompt giving us the option leads me to assume that I still have Nuttx as a choice of OS.
Thank you,
Ivan
Posts: 1
Participants: 1
@chobitsfan wrote:
Hi @priseborough @peterbarker I am a little confuse about ek2_flow_delay or ek2_XXX_delay. It means the sensor processing latency? ek2_flow_delay means the latency from flow sensor capture an image to it finished processing and output result? If sensor sampling rate increased from 10hz to 20hz, will it effect ek2_flow_delay? For example, my camera fps increased from 10fps to 20fps, but my mcu didn’t change and the processing latency remains the same. Thank you very much
Posts: 1
Participants: 1
@smilie wrote:
I change from px4+nuttx to arduplane+chibios (3.9.8?)
Now i have no power on my vdd_5v_periph_en line for powering the gps-port and i2c-port.
I have no idea, how can i get the normal function back.
thanks
Posts: 1
Participants: 1
@lucamax wrote:
Arducopter 3.5.5
I have recently replace the original 3DR Gps with a new one from Mrobotics, Neo-M8N Dual Compass LIS3MDL+ IST8310, but find out the the two compass do not work with the original Pixhawk version 2.4 .
First I tried to install a Zubax Gps but for unknown reasons the barometer and the compass did not work.So finally I install again the Mrobotics Neo-M8N with an external compass, the well known MHM5983 .
GPS work fine, compass calibration perfect.
It flies very well .The problem: At each boot I receive a message asking for a new 3D accell calibration
Another weird thing is that before taking off , the status led is green , when I start the motors it goes red and yellow, a message appears on MP saying , something like “switching to EKF 1” then the led returns green.
I do not know if it is related with 3D accell calibration request.
Posts: 1
Participants: 1
@Dorian wrote:
Hello,
I am interested in the Herelink system.
I would like to know if there is a solution to connect a camera with an analog output to Herelink Air Unit?
Posts: 1
Participants: 1
@YuriuSS wrote:
I have an issue in Pos.Hold mode
Octorotor x, Pixhawk Cube2.1 + HERE2, AC3.6.8 after takeoff or first time turning on Pos.Hold mode turn on in the air, the drone is swinging with big angles like it has a big problem with compass, but after few movements right-left, forward-backward it flying very precise and good, automatic mission, RTL, holding position - all working correct. Just has a problem in first 30-60 sec after takeoff.AHRS = 0, Compass orientation = 0,
In Mission Planner all compasses are match and showing correct orientation all the timeSame problem in two very different octa drones with different gps\compass modules
Posts: 1
Participants: 1
@Vandan_Rao wrote:
Hi All,
The UC4H modules by OlliW have done some great progress into the development of UAVCAN into Ardupilot.
I am working on a parallel to slowly convert all sensors and actuators of the drone into CAN compatible. I’d love to see one single CAN bus connecting all sensors and actuators to the Cube.I feel that the SERIALTNL params of OlliW’s Betacopter can be really useful in general if they are included in the params list of ArduPilot.
Any particular reason they were not added? I’d like to actually merge the params cuz that might give us some tunnelling for UARTs through CAN1/CAN2. Has this been achieved through any other means?
Posts: 2
Participants: 2
@flightzoomer wrote:
These days the GCS FlightZoomer 3.2 has been released which offers a number of new features:
- Full stop ILS landings for copter.
- Support for Sky Viper.
- Enhanced ROI handling with staged ROI’s.
- Predefined sweep times for gentle movements from one ROI to the next.
- FlightZoomer’s own 3G telemetry depreciated. Support for Andruav for 3G telemetry instead.
- And the most outstanding feature: ROI sequences
ROI sequences work like this
- Like missions have waypoints, a ROI sequence has ROI’s. The camera pans in one steady movement from the first ROI to the last ROI.
- The trick is, that flown missions (flightplans in FlightZoomer terminology) are executed in synchronization with the ROI sequence. At each waypoint the camera will be pointing to the next ROI in the sequence!
- ROI sequences can be planned and simulated with the in-built simulation feature (using the huge synthetic FPV screen).
- ROI sequences can be stored so they can easily be repeated in the future.
- As an alternative to the waypoint synchronization, ROI sequences can also be defined time based. For time-based ROI sequences, the sweep time in seconds is simply defined per ROI.
- Get an impression in the video above, how precise the same footage can be repeated and how seamless you can blend between different shots of the same ROI sequence.
- Documentation of ROI sequences
Staged ROI’s work like this
- Up to four ROI’s can be pre-defined as staged ROI’s
- The user can trigger flexible camera movements by activating one of the staged ROI’s at any time.
- The duration of the sweep to the triggered staged ROI can be defined by the user.
- Additionally, an altitude offset can be applied to any ROI, which is targeted for a sweep.
- The two parameter sweep time and altitude offset can simply be reused or freshly defined prior any sweep.
- Documentation of staged ROI’s
Copter full stop ILS landings work like this
- After the glideslope has been captured, the copter descends on the final approach down to 7 Meter above runway elevation.
- At that point, the descend stops and the copter continues to the runway at constant altitude.
- Precisely over the runway threshold, the horizontal movement stops and ArduCopters LAND mode is activated (before FlightZoomer controlled the flight path using GUIDED mode).
- Documentation of FlightZoomer ILS approaches
New pricing
A significant change in the price model makes FlightZoomer much more attractive: the full functional scope as a GCS up to version 3.1 is now free again. Only selected new features will be added as in-app purchases. Up to now, the only feature that has a price are ROI sequences. Even using the staged ROI’s is free.
Some links:
Posts: 1
Participants: 1
@MastaMinds wrote:
Hello ArduPilot Community!
I am trying to build ArduCopter firmware in a Linux environment, but I am having some issues at the final step (the compilation step).I followed all the steps listed in:
http://ardupilot.org/dev/docs/building-setup-linux.html
(Got a little confused because I was following a Youtube video, and discovered that apparently, the steps change and are updated on an hourly basis!)I was successful in most of the steps. However, when I reached the last step of compiling the code, it begins without issues and then stops when compiling
“libraries/AC_AttitudeControl/AC_AttitudeControl_Sub.cpp”Not just that, but also a bunch of other files.
And the error I get is as follows[7/8] Creating build/Pixhawk1/ap_version.h
[ 9/552] Compiling libraries/AC_AttitudeControl/AC_AttitudeControl_Sub.cpp
[ 10/552] Compiling libraries/AC_AttitudeControl/AC_AttitudeControl_Heli.cpp
[ 11/552] Compiling libraries/AC_AttitudeControl/ControlMonitor.cpp
[ 12/552] Compiling libraries/AC_AttitudeControl/AC_PosControl_Sub.cpp
In file included from …/…/libraries/AP_Param/AP_Param.h:25:0,
from …/…/libraries/AC_AttitudeControl/AC_AttitudeControl.h:7,
from …/…/libraries/AC_AttitudeControl/AC_AttitudeControl_Sub.h:6,
from …/…/libraries/AC_AttitudeControl/AC_AttitudeControl_Sub.cpp:1:
…/…/libraries/AP_Common/missing/cmath:3:22: fatal error: cmath: No such file or directory
#include_next
^
compilation terminated.I tried removing gcc compiler, gcc-arm-none-eabi, and tried getting other versions. But still useless. I tried 2017, 2018 and in the end, I tried the 2015 version (because it is the one listed in the Wiki), but still with no success.
What do you think the problem is?Thank you!
Posts: 1
Participants: 1
@Jose_Ortiz wrote:
Hello I have a cube black on a small quadcopter that I have fully configured to fly but I keep getting a message that isn’t allowing the copter to arm. I have done the recommended changes as the service bulletin suggested, have latest mission planner installed, run 3.6.8 cube black ardupilot and set the parameters according to the post.
I am running the gps through the GPS1 port so that the safety switch and everything else still functions properly.
I have already reseated the cable connection from the inside the gps and on the carrier board.
And the small switch has been toggled and made sure to be set to I2C. I was wondering if anyone else has run into this issue and any guidance on how to fix it.
Thank you.
params with gps config issue.param (14.3 KB)
Posts: 4
Participants: 2
@spencer1 wrote:
I am hoping to use Arduino specific functions in the UserCode.cpp file of ArduCopter. I specifically want to use the functions in the SPI library. Also, I am using the Pixhawk flight controller.
Are these already included in the ArduCopter source code? Or do I need to insert them into header files and #include them to use them?
Posts: 1
Participants: 1
@GFChr wrote:
I am sure it is something simple I am somehow overlooking.
Vehicle has previously operated fine, with many launches in alt hold, and loiter mode.
Now, any launch in either mode causes an instant ground loop (pitch direction, Nose up and rolling backwards every time).
Vehicle launches fine in stabilize, and doesn’t shift position at all when converting the airborne vehicle from stabilize to alt hold or loiter.
Somehow I am sure this is an operator induced problem, just can’t resist frolicking about in the full parameter list. Any hints of where to begin unraveling problem would be greatly appreciated.
Posts: 1
Participants: 1
@luckyflyer wrote:
I have a new Pixhack V3 and I’m trying to decide which heli to put it on. I have a logo 480 a logo 550 and a M.A. whiplash 700 gasser.
In my youth I would check the depth of the waters by just jumping in, I’m a little older now and want to check with one foot or even toe. I’m simply saying that I would like to put the Pixhack on the 700 gasser but for the first time due to crash costs I’m leaning towards the 480. But even with lower head speeds and the largest possible battery I realize flight times are limited compared to a gasser.
I have no real purpose or mission for uav copter other than to try one and have a new toy. I have several goggles from the fatshark hdo’s down. Do well balanced heli’s provide a stable image? Mission planner or something else?
These and probably many other questions that need to be answered before I pull a Neo out of one of these helicopters.
Posts: 1
Participants: 1
@Bruce_Baxter wrote:
Help,
i have been using mission planner and apm 2.xx and pixhawk 2 for many years with no problem, but …there seems to be a driver clash when i try to enable the joystick in mission planner. It briefly recognises what joystick is connected and 3-5 seconds later mission planner crashes and closes.
I have tried while connected to pixhawk, not connected to pixhawk, enter configuration before plugging in joystic, joystic plugged in first.
all situations crash 3-5 seconds after enabling joystick.thankyou for your help on this matter
Posts: 2
Participants: 1
@talos wrote:
Hi
I’d like to send data such as rpm using serial port following the instruction,
http://ardupilot.org/dev/docs/learning-ardupilot-uarts-and-the-console.html
but it did not work like this，
hal.console->printf(rpm)；
Which serial port should I use, did I need to set the serialx_protocol? Thanks
I do not like to use mavlink
Posts: 1
Participants: 1
@birdmanpete wrote:
I see others have asked this question but I cannot see the solution. The Mission Planner setup checks out on every screen but the mode tally sheet for channels 5 and six do which work precisely on the radio calibration page do not respond as predicted on the modes page where they cannot be saved!
Have I got a dud Pixhawk?
Or should I reinstall Mission Planner.Birdmanpete.
Posts: 1
Participants: 1
@MastaMinds wrote:
Hello Ardupilot community!
I am trying to build the ArduCopter firmware, but I am having some issues. At first, I was following a Youtube video, but I then realized that things have changed a bit, and I followed up the steps listed on the website, this link:
http://ardupilot.org/dev/docs/building-setup-linux.html
(I am using Ubuntu 18.10).Most of the steps were successful, however, I am facing some issues. At first, I didn’t know which board to select… Pixhawk1 or pxf? (I am assuming pxf was px4-v2 previously).
I decided to configure waf for Pixhawk1 (I am using the original Pixhawk with px4 installed).
But when I first tried to build the firmware I got this error:[ 9/552] Compiling libraries/AC_AttitudeControl/AC_AttitudeControl_Sub.cpp
[ 10/552] Compiling libraries/AC_AttitudeControl/ControlMonitor.cpp
[ 11/552] Compiling libraries/AC_AttitudeControl/AC_AttitudeControl_Heli.cpp
[ 12/552] Compiling libraries/AC_AttitudeControl/AC_PosControl_Sub.cpp
In file included from …/…/libraries/AP_Param/AP_Param.h:25:0,
from …/…/libraries/AC_AttitudeControl/AC_AttitudeControl.h:7,
from …/…/libraries/AC_AttitudeControl/ControlMonitor.cpp:1:
…/…/libraries/AP_Common/missing/cmath:3:22: fatal error: cmath: No such file or directory
#include_next
^
compilation terminated.In file included from …/…/libraries/AP_Param/AP_Param.h:25:0,
from …/…/libraries/AC_AttitudeControl/AC_AttitudeControl.h:7,
from …/…/libraries/AC_AttitudeControl/AC_AttitudeControl_Sub.h:6,
from …/…/libraries/AC_AttitudeControl/AC_AttitudeControl_Sub.cpp:1:
…/…/libraries/AP_Common/missing/cmath:3:22: fatal error: cmath: No such file or directory
#include_next
^
compilation terminated.In file included from …/…/libraries/AP_Param/AP_Param.h:25:0,
from …/…/libraries/AC_AttitudeControl/AC_PosControl.h:4,
from …/…/libraries/AC_AttitudeControl/AC_PosControl_Sub.h:3,
from …/…/libraries/AC_AttitudeControl/AC_PosControl_Sub.cpp:1:
…/…/libraries/AP_Common/missing/cmath:3:22: fatal error: cmath: No such file or directory
#include_next
^
compilation terminated.In file included from …/…/libraries/AP_Param/AP_Param.h:25:0,
from …/…/libraries/AC_AttitudeControl/AC_AttitudeControl.h:7,
from …/…/libraries/AC_AttitudeControl/AC_AttitudeControl_Heli.h:6,
from …/…/libraries/AC_AttitudeControl/AC_AttitudeControl_Heli.cpp:1:
…/…/libraries/AP_Common/missing/cmath:3:22: fatal error: cmath: No such file or directory
#include_next
^
compilation terminated.
Waf: Leaving directory `/home/osboxes/ardupilot/build/Pixhawk1’
Build failed
-> task in ‘objs/AC_AttitudeControl’ failed (exit status 1):
{task 139683473623712: cxx ControlMonitor.cpp -> ControlMonitor.cpp.0.o}
(run with -v to display more information)
-> task in ‘objs/AC_AttitudeControl’ failed (exit status 1):
{task 139683473623472: cxx AC_AttitudeControl_Sub.cpp -> AC_AttitudeControl_Sub.cpp.0.o}
(run with -v to display more information)
-> task in ‘objs/AC_AttitudeControl’ failed (exit status 1):
{task 139683473623592: cxx AC_PosControl_Sub.cpp -> AC_PosControl_Sub.cpp.0.o}
(run with -v to display more information)
-> task in ‘objs/AC_AttitudeControl’ failed (exit status 1):
{task 139683473623352: cxx AC_AttitudeControl_Heli.cpp -> AC_AttitudeControl_Heli.cpp.0.o}
(run with -v to display more information)I tried removing and re-installing gcc-arm-none-eaib, but with no use, I tried to use the compiler in the website, still no use, I installed and removed cmake, g++, and still all efforts are in vain… And nothing changed. I am pretty sure the issue is something with the compiler, but I don’t know where exactly.
I tried another path, which is to configure the waf to change the borad to pxf, and this is the result I got:
Setting top to : /home/osboxes/ardupilot
Setting out to : /home/osboxes/ardupilot/build
Autoconfiguration : enabled
GCOV code coverage analysis : no
Setting board to : pxf
Using toolchain : arm-linux-gnueabihf
Checking for ‘g++’ (C++ compiler) : not found
Checking for ‘clang++’ (C++ compiler) : not found
could not configure a C++ compiler!
(complete log in /home/osboxes/ardupilot/build/config.log)Which means it cannot even find a C++ compiler, which doesn’t make senes because I have already installed them. (I also did the step of adding the bin file of gcc-none-arm-eabi to my directory, but it’s all the same)
I would be grateful if you could help me!
Thank you!Edit 1: The build command works fine for SITL.
Posts: 1
Participants: 1
