@tab8429 wrote:
(topic withdrawn by author, will be automatically deleted in 24 hours unless flagged)
Posts: 1
Participants: 1
@tab8429 wrote:
(topic withdrawn by author, will be automatically deleted in 24 hours unless flagged)
Posts: 1
Participants: 1
@juibn wrote:
Hello.
i`m testing Object avoidance Bendy ruler.
when i first test, Using copter 4.0.2
Pixhawk cube Black
RPLIDAR A3when set OA - 1, and start mission, first Waypoint is working.
not continue next waypoint. jusy stuck in air. just hovering.so, i`m using orange cube and SF40/C.
Same @rmackay9 Youtube video setting.copter v4.0.3
i`m test today again.
when first and second test, it is good working. but after same situation is occur.
not continue next waypoint.Here is my Log and Parameter.
OA_test.param (17.9 KB)
How can i solve this problem ?
Thank you.
Posts: 1
Participants: 1
@felis wrote:
Hello. I’m having an issue with getting the Pozyx works for the ArduRover.
I’ve followed all the steps in https://ardupilot.org/copter/docs/common-pozyx.html. I don’t need the Z coordinate, so I configure the IndoorLoiter to run 2.5 D positioning.
The global position latitude & longitude data are always 0. It never jump to the BCN_LATITUDE and BCN_LONGITUDE as I expected. Pozyx serial output looks fine though. I’ve also checked the DataFlash Log and all the BCN data looks normal.
I noticed that I always get EKFCHECK-2. Based on what I read, it means there are at least 2 variance errors on the EKF. I initially tried EKF2, then I switched to EKF3, but the result is the same. Here are the logs and parameters:
https://drive.google.com/drive/folders/1vu8FVQcEdwxcMqs6EvUGBe2JvrnX4vfKI am not sure whether the error is caused by wrong parameter settings or hardware setup. FYI, I do not use compass or GPS.
Do you have any suggestions? Thanks.
Posts: 1
Participants: 1
@111194 wrote:
Hello,
We want to use auto mode indoors without GPS using beacon of Marvelmind.
We followed this manual.
So we got 3D FIX. But at the same time, the EKF turns red. I don’t know if this is normal.
The main problem is that the propeller doesn’t spin.
When i try to arming, the phrase ‘Arm: Compass not healthy’ appears.
But the motor has a melody when it is arming. However, the motor does not spin when throttle is raised up. And the screen is also DISARMED.
Can I get some help on this?
Thank you…
Posts: 1
Participants: 1
@Mert_Askin wrote:
Hi,
I’m using pixhawk 2.4.3 and copter-4.0.3 version installed on.
I tried to control the copper with attitude control using dronekit API.(set_attitude_target)
I used this function for this;def send_attitude_target(self, roll_angle=0.0, pitch_angle=0.0, yaw_angle=None, yaw_rate=0.0, use_yaw_rate=False,thrust=0.5): if yaw_angle is None: # this value may be unused by the vehicle, depending on use_yaw_rate yaw_angle = self.vehicle.attitude.yaw # Thrust > 0.5: Ascend # Thrust == 0.5: Hold the altitude # Thrust < 0.5: Descend msg = self.vehicle.message_factory.set_attitude_target_encode( 0, # time_boot_ms 1, # Target system 1, # Target component 0b00000000 if use_yaw_rate else 0b00000100, self.to_quaternion(roll_angle, pitch_angle, yaw_angle), # Quaternion 0, # Body roll rate in radian 0, # Body pitch rate in radian math.radians(yaw_rate), # Body yaw rate in radian/second thrust # Thrust ) self.vehicle.send_mavlink(msg)
Copter is in guided mode while atitude control command is running.When I do any intervention from remote controller(RC), the code changes the copter to rtl mode.I assigned stabilize and land mode to the stick on the remote controller.
2020-03-25 14-14-31.bin (969.7 KB)
First Flight:
- Atitude control command worked. The copter was switched land mode from remote control(RC)to bring it back.
- It can be seen from the logs. It changes from land mode to rtl mode.
- As a result, the copter came back to where it took off and was disarmed
Second Flight:
- Atitude control command worked. The copter was switched stabilize mode from remote control(RC)to bring it back.
- It can be seen from the logs. It changes from stabilze mode to rtl mode.Meanwhile, the throttle stick on the remote controller was at its lowest position.
- Although it is seen that rtl mode in the logs, the copter has fallen to the ground and is broken
What could be the reason for this?Can mod change cause this?
Thanks.
Posts: 6
Participants: 2
@Ravi_Singh wrote:
Hello Guys,
I know this may not be a right forum with the query I am about to ask. But, I guess a lot of people in the group who are into aerial mapping and are quite experienced as well. So I may get some help here.
I had a query regarding using sony rx0 camera for mapping application as its compact, rugged and lightweight platform. However, it lags Hotshoe feedback… Lots of people have already have ruled out the possibility of feedback from this camera. But there is one company https://dronee.aero/pages/dronee-plane-wing-rtk-ppk which claims that they can do that.
Is it even possible with this camera?
Is it just a hoax?
How is it possible to do it? Is there a sizable number of people on this forum who have done it and have got positive results?
Posts: 2
Participants: 2
@Maurice wrote:
I wanted to use the example code for roll LEDs, but there are some parts where this doesn’t seem to work for me. Here is the relevant part of the code:
local update
local num_leds = 9–[[
use SERVOn_FUNCTION 94 for LED. We can control up to 16 separate strips of LEDs
by putting them on different channels
–]]
local chan = SRV_Channels:find_channel(94)if not chan then
gcs:send_text(6, “LEDs: channel not set”)
return
end– find_channel returns 0 to 15, convert to 1 to 16
chan = chan + 1gcs:send_text(6, “LEDs: chan=” … tostring(chan))
– initialisation code
–serialLED:set_num_neopixel(chan, num_leds)
–serialLED:set_num_profiled(chan, num_leds)serialLED:set_num_LEDs(chan, num_leds)
serialLED:set_RGB(chan,1,0,0,0)
the 2 functions set_num_neopixel and set_mum_profiled can’t be found and I can’t find them in any doc. So I switched to set_num_LEDs…
when running the last line in the code i get the error message bad argument #2 in QGroundControlCan somebody explain what this means, and what could be the reasons. I tested so much and still couldn’t figure it out…
Posts: 2
Participants: 2
@anbello wrote:
MAVCONN library (libmavconn) is the mavlink connection and communication library used in MAVROS that could be used also outside the ROS environment.
I used it in two project related to non-GPS Navigation systems based on Visual Position Estimation and I would like to share with Ardupilot community my experience.
The first project (aruco_mavconn) is the porting to (non-ROS) C/C++ of the positioning system based on Aruco Boards described in this wiki page (https://ardupilot.org/dev/docs/ros-aruco-detection.html).
The intent was that of learning how to use libmavconn and a little bit more of OpenCV while at the same time obtaining a lighter application to run on Single Board Computer.
The second project (t265_mavconn) is the porting to C/C++ of the Python script written by Thien Nguyen (@LuckyBird) to send VISION_POSITION_ESTIMATE messages from the T265 VIO device to Flight Controller after transforming the reference frame to NED. The work done by Thien Nguyen is documented in this wiki page (https://ardupilot.org/copter/docs/common-vio-tracking-camera.html).
Other then porting to (non-ROS) C/C++ this two systems I added the possibility to choose between two way of sending the Visual Position Estimation messages:
- using EKF2 + sending VISION_POSITION_ESTIMATE message;
- using EKF3 + sending GPS_INPUT message.
The use of GPS_INPUT message permit us to send information about velocity other then position and attitude. In adding this feature I used as reference the Vicon MAVProxy module written by @Tridge for the Vicon Indoor Positioning System documented here. From this module I used the part relative to time and coordinate transformation (from local NED to GPS).
Both projects has been tested on Copters with Raspberry Pi 3 (with Ubuntu 16.04) used as Companion Computer. aruco_mavconn has been tested also on a desktop PC with Ubuntu 18.04 using a small (3 inch) Copter with Raspberry Pi Zero W connected in WiFi to The PC. The configuration relevant to this last system is explained in this post (Indoor autonomous flight with Arducopter, ROS and Aruco Boards Detection).
For the connection of the Companion Computer to Flight Controller and relative settings see relevant information on the wiki pages:
Intel RealSense T265 to use T265 VIO Camera with t265_mavconn;
ROS with Aruco Boards detection to use Aruco Boards with aruco_mavconn.
I have to say that I am not so good at C++ object oriented programming, so the programs consist mainly of one source file with all functions and logic in there. I hope to find the time to refactor the sources in a better form.
aruco_mavconn
aruco_mavconn depends on OpenCV and MAVCONN.
To build OpenCV form source follow the instruction at:
https://docs.opencv.org/master/d7/d9f/tutorial_linux_install.html
To build MAVCONN follow the instruction in the relative paragraph of this post.
After installing the prerequisites it is possible to built the project in this way:
$ git clone https://github.com/anbello/aruco_mavconn $ cd aruco_mavconn $ mkdir build $ cd build $ cmake .. $ make $ cp ../data/* .
To start the program:
$ detect_board_mavconn –lp=../data/launch.yml
Here an example of launch file (lines beginning with ‘#’ are ignored):
%YAML:1.0 dictionary: 16 camera_param: "calib-raspicam2.3.yml" #video_pipeline: "udpsrc port=9000 ! application/x-rtp,media=video,clock-rate=90000,encoding-name=JPEG,a-framerate=30,payload=26 ! rtpjpegdepay ! decodebin ! videoconvert ! appsink" video_pipeline: "" camera_id: 0 offset_x: 0.0 offset_y: 0.05 offset_z: 0.0 board_layout: "layout-my.yml" detector_param: "detector_params.yml" refined_strategy: 0 show_rejected: 0 mavconn_url: "tcp://192.168.10.16:2000" #mavconn_url: "/dev/ttyACM0:115200" vision_gps_msg: 2 # 0 -> no msg | 1 -> vision msg | 2 -> gps msg pose_msg_rate: 15 show_image: 1
Video source can come from a gstreamer pipeline (if video_ pipeline string is not null) or v4l2 device.
In the data directory there are some example of camera calibration file, board layout file and detector parameter file.
t265_mavconn
t265_mavconn depends on librealsense, OpenCV and MAVCONN.
To build librealsense follow the instruction at:
https://github.com/IntelRealSense/librealsense/blob/master/doc/installation.md
To build OpenCV form source follow the instruction at:
https://docs.opencv.org/master/d7/d9f/tutorial_linux_install.htmlTo build MAVCONN follow the instruction in the relative paragraph of this post.
After installing the prerequisites it is possible to built the project in this way:
$ git clone https://github.com/anbello/t265_mavconn $ cd t265_mavconn $ mkdir build $ cd build $ cmake .. $ make $ cp ../data/* .
To start the program:
$ t265_mavconn –lp=launch.yml
Here an example of launch file (lines beginning with ‘#’ are ignored):
%YAML:1.0 camera_orientation: 0 pose_msg_rate: 15 vision_gps_msg: 1 # 0 -> no msg | 1 -> vision msg | 2 -> gps msg offset_x: 0.10 offset_y: 0.0 offset_z: 0.0 scale_factor: 1.0 #mavconn_url: "tcp://192.168.10.16:2000" mavconn_url: "/dev/ttyACM0:115200"
Useful parameters
On both programs the parameter
vision_gps_msg
(in launch file) permit to choose betweenVISION_POSITION_ESTIMATE
(1) messages orGPS_INPUT
(2) messages. Withvision_gps_msg: 2
both pose and velocity are sent to ArduCopter. In aruco_mavconn the velocity is calculated from actual and previous position, in t265_mavconn the velocity come from the T265 device using librealsense.When using VISION_POSITION_ESTIMATE message the following parameter should be set in ArduCopter:
AHRS_EKF_TYPE = 2 EK2_ENABLE = 1 EK3_ENABLE = 0 EK2_GPS_TYPE = 3 EK2_POSNE_M_NSE = 0.1 EK2_VELD_M_NSE = 0.1 EK2_VELNE_M_NSE = 0.1 GPS_TYPE = 0 COMPASS_USE = 0 COMPASS_USE2 = 0 COMPASS_USE3 = 0
When using GPS_INPUT message the following parameter should be set in ArduCopter:
AHRS_EKF_TYPE = 3 EK2_ENABLE = 0 EK3_ENABLE = 1 EK3_ALT_SOURCE = 2 EK3_GPS_TYPE = 0 EK3_MAG_CAL = 5 EK3_POSNE_M_NSE = 0.1 EK3_VELD_M_NSE = 0.1 EK3_VELNE_M_NSE = 0.1 GPS_TYPE = 14 GPS_DELAY_MS = 50 COMPASS_USE = 0 COMPASS_USE2 = 0 COMPASS_USE3 = 0
Building MAVCONN library on Linux (Ubuntu 16.04 / 18.04)
$ sudo apt-get install catkin libconsole-bridge-dev libboost-dev libboost-system-dev python-dev python3-dev python-future $ git clone https://github.com/mavlink/mavlink-gbp-release.git $ cd mavlink-gbp-release $ git checkout release/melodic/mavlink $ mkdir build $ cd build $ cmake .. $ make $ cd ../.. $ export mavlink_DIR=./mavlink-gbp-release/build $ git clone https://github.com/mavlink/mavros $ cp -r mavros/libmavconn . $ cd libmavconn $ mkdir build $ cd build $ cmake .. $ make $ sudo make install $ cd ../.. $ sudo mkdir /usr/local/include/mavlink $ sudo cp -r mavlink-gbp-release/build/include/v2.0/ /usr/local/include/mavlink
Posts: 1
Participants: 1
@Koenvrints1 wrote:
I’m looking for the 3D file (step) of the Here/here+/here2 GPS someone has this file?
Posts: 1
Participants: 1
@LucasC wrote:
Hi everyone!
Thanks for reading this.
I’m creating a product for my school, and I’m using the Pixhawk 2.4.8 flight controller board. I have Mission Planner and I have used it to calibrate the accelerometer and a few other things.
My problem is that all the channels, such as pitch, yaw, etc are working properly and I’m getting feedback from Mission Planner showing the inputs from my transmitter. However only channel 1 is not working. I am using a FlySky 10 channel Rx and the PPM encoder that came in the HobbyPower S550 kit.
One thing I noticed is that channel 1 shares a plug with the positive and ground pins. Not sure if that means anything, but it’s an observation.
Does anyone know about this issue?
Thanks everyone for your time!
-Lucas C
Posts: 1
Participants: 1
@Ludo41 wrote:
Hello,
My log file does not have the correct date in the file name.
For example, today the file should be called 2020-03-27 then the time, may if it is called 2020-03-24 then the time that it is correct.
My GPS sends the NMEA GGA, RMC and VTG frames, the RMC has the correct date.
My Windows system has the correct date.
Do you have any ideas?
Thanks
Ludo41
Posts: 1
Participants: 1
@Mustafa_Gokce wrote:
Hi.
Is there a way to force the copter to before navigating to home location, adjust the altitude to desired return to launch altitude.
I know that if the copter’s current altitude is below the RTL_ALT, the copter climbs to RTL_ALT before navigating to RTL location but is there a mechanism to descent to RTL_ALT if i am above RTL_ALT before going to return to launch position?
If not how can i do this?
Thanks…
Posts: 2
Participants: 2
@dkemxr wrote:
I have an issue with a 500 size quad that I’m not sure how to attack. Tuning is good, vibes are OK (small amount of clip at fast speed). At hover all is well but with horizontal speed the motors are being commanded to oscillate.with a fairly high range of magnitude. This craft actually flies well but this is bothering me. Craft description:
Pixracer
1.7kg AUW
3508-380kv Motors
Some older BlHeli_S ESC’s
13" CF props
6S PowerTake a look at the FFT and the peak at 29Hz (post filter). I’m using the Notch filter to handle a 59Hz mode and it’s harmonics perhaps I should address this 29Hz mode?
https://www.dropbox.com/s/zjwgi3hvikn8o8m/500%20quad.bin?dl=0
Posts: 3
Participants: 2
@Supersurfer wrote:
Hi, I hope somebody can help me.
I have flown with a pixhawk 2: black cube for some time with APM:Copter 3.5.0 RC4.
Everything was working fine until I upgraded to the latest ArduCopter V4.0.3.
If I take off in Loiter or Stabilise and keep the quad at a hight of a few meters it is not staying steady and crashes by falling to one side (it does not stay leveled). can this be a difference in the newer software? Nothing else has been changed.
Posts: 2
Participants: 2
@mbuilds wrote:
Hello
I recently flew a short mission, The mission appeared near perfect low vibe ect and I decided to allow the copter (quad Pixhawk 2.1 ) kde esc and motors) lad in auto after the mission completed… I had "autotuned the copter a month or more prior and never checked it for ossolations on RTL… the copter looked great until…
It touched down but the motors did not disarm instead they continued and after a few minutes the drone tilded backward causing the 2 rear motors to stop as they hit the ground… I then helt the copter down to prevent any damage… I was hoping someone could look at the log and give me insight as to why this happened and assure me it will be of to fly (I ran a motor test since and all motors spin up correctly and in the correct direction… log is at link below
https://1drv.ms/u/s!AmW8JONezY1Ugftcf-42hntC4hTjGg?e=xZWgUzthanks for support
Mike
Posts: 1
Participants: 1
@fpv_duck wrote:
Hi,
I’m new to ardupilot and UAVs/ROVs (4 months now), and I’m not sure how I’m supposed to handle this behavior:
I have managed to make a pseudo-working rover. Here’s my minimal setup:
- Pixhawk 2.4.8 clone
- Brushed ESC (but behaves as a brushless ESC, I can’t get it to work with MOT_PWM_TYPE other than “Normal”)
- I’m connected to qgroundcontrol on windows using an USB cable
- I’m trying to control the rover using an usb gamepad
My goal is to control the rover using UAVMatrix in 4g, which is why I’m not using a traditional RC transmitter.
My problem is that in manual mode, my gamepad joystick can change the SERVO3 value from SERVO3_MIN (which I set to 1000), and SERVO_3 TRIM (which I set to 2000).
Now using this setup it works. When the gamepad throttle is in the middle I have PWM at 1500 and the vehicle is idle.
The problem is that when it switches to “Hold” mode, then SERVO3 value goes to 2000 (because it’s TRIM, that’s expected), and it goes full throttle.
Also I had to disable the PWM output when disarmed, because if not then disarming the vehicle as the same effect. And that’s a bit annoying because when I arm the vehicle it goes full throttle for half a second (probably because the ESC catches up).
So I’m wondering why the gamepad doesn’t allow the values to go from SERVO3_MIN to SERVO3_MAX?
Here’s my parameters in case that helps: roverpb.txt (26.4 KB)
Let me know if I can provide you any other information.
Thanks for your help!
Posts: 4
Participants: 2
@ninja_zx11 wrote:
Hi all,
I just finished installing arducopter on my Blade 450 3D with a flybar.Flight controller is omnibusF4V3Pro and everything is working as it should including throttle interlock,GPS etc.Swash plate and all controls are moving as per the wiki.This blade 450 was flying fine prior to switching over to arducopter.So swash plate is levelled,linkages swash plate movements were all set up right before.
Now i didn’t understand these settings properly:
“Setting Collective Pitch Range and Zero Thrust Point”
" Setting Maximum Cyclic Pitch"
When i flip the interlock switch,motor starts spinning at a constant speed.Is it a normal behavior?I have not set up any governor.Esc is castle creations Talon 35A which has heli functions but i have not touched its settings yet.Before i install my rotor blades,i would like to get suggestion from you guys.I am also thinking to install training gear just for safety during a maiden. Thanks.
Posts: 1
Participants: 1
@Autonomost wrote:
Hi All,
I am seeing a lot of bad packets on mission planner when connecting via UDP. Can anyone decipher what mission planner is complaining about?
ssionPlanner.MAVLinkInterface - MAVLINK: bad read 6, 36, 5
INFO MissionPlanner.MAVLinkInterface - Mavlink Bad Packet (crc fail) len 36 crc 31471 vs 0 pkno 163 AHRS
Unknown Packet
INFO MissionPlanner.MAVLinkInterface - unknown packet type 58
INFO MissionPlanner.MAVLinkInterface - mav 1-1 seqno 47 exp 44 pkts lost 3
INFO MissionPlanner.MAVLinkInterface - MAVLINK: bad read 10, 40, 9
INFO MissionPlanner.MAVLinkInterface - Mavlink Bad Packet (crc fail) len 40 crc 63157 vs 0 pkno 30 ATTITUDE
INFO MissionPlanner.MAVLinkInterface - MAVLINK: bad read 6, 40, 5
INFO MissionPlanner.MAVLinkInterface - Mavlink Bad Packet (crc fail) len 40 crc 42736 vs 0 pkno 241 VIBRATION
=@@{+"=>oINFO MissionPlanner.MAVLinkInterface - mav 2-1 seqno 0 exp 255 pkts lost 1
INFO MissionPlanner.MAVLinkInterface - MAVLINK: bad read 10, 48, 9
INFO MissionPlanner.MAVLinkInterface - Mavlink Bad Packet (crc fail) len 48 crc 7829 vs 0 pkno 147 BATTERY_STATUS
dINFO MissionPlanner.MAVLinkInterface -
@==sDF<fA5sM71Df6Vd}m9)R<:X:ud:5R@=f(IJ=@@{+"=>odINFO MissionPlanner.MAVLinkInterface - mav 2-1 seqno 17 exp 16 pkts lost 1
INFO MissionPlanner.MAVLinkInterface - MAVLINK: bad read 6, 36, 5
INFO MissionPlanner.MAVLinkInterface - Mavlink Bad Packet (crc fail) len 36 crc 6768 vs 0 pkno 30 ATTITUDE
@Q?INFO MissionPlanner.MAVLinkInterface - mav 1-1 seqno 51 exp 49 pkts lost 2
INFO MissionPlanner.MAVLinkInterface - MAVLINK: bad read 6, 30, 5
INFO MissionPlanner.MAVLinkInterface - Mavlink Bad Packet (crc fail) len 30 crc 58033 vs 0 pkno 116 SCALED_IMU2
VUnknown Packet
[INFO MissionPlanner.MAVLinkInterface - mav 1-1 seqno 56 exp 54 pkts lost 2I am trying to connect to two aircraft over UDP through one groundstation.
Any info would be great! Thanks
Posts: 1
Participants: 1
@LeeM wrote:
Hi,
I’m just putting together a ground station and I need a little advice on the best way of getting my A/V video in to the window in AMP2.0 where the flight data is.
I know how to select the video as I can select my webcam, but I need a suggestion on some proven A/V to USB capture device and some guidance on how to do it.Thanks
Posts: 1
Participants: 1
@diego_supairvision wrote:
Hi everyone!
So today, after doing a regular flight, I have tried to update my quad to the latest version (from 4.01 to 4.03).
The installation (using mission planner) has not been problematic except at the end. When it says “wait for the tune before OK” I’ve waited for around 2 minutes and the buzzer has not made its regular sound. I’ve thought that the buzzer might have sounded earlier without me noticing it so I`ve clicked OK and proceed to reboot the Cube.
So now, when I power the cube (USB or battery) the buzzer does not do the initial sound and in mission planner it says “Check BRD_TYPE: Failed to init CubeBlack - sensor”. BRD_TYPE is se to 3. I have tried to reboot it multiple times, flash older firmwares and nothing.
I’ve checked the HW ID page and this is what it shows. Isn`t the barometer missing?
It is a pixhawk Cube Black that I bought nearly a year ago. I have updated the firmware multiple times with the same computer and I have never had a problem. Anyone knows the cause and how can I solve it?
Thanks in advance!
Posts: 10
Participants: 2