This is an impressively stable test. Did you use the old firmware (px4flow-klt-06Dec2014) for this test flight or did you use a newer version from the PX4 repo? Cheers!
Hi everyone! I'm testing v3.3.3 using PIXHAWK and PX4FLOW. I read reply Goro Senzai and all comments. And quadrotor work is realy cool! But I still have question, when trying fly with coaxial octorotor. Octorotor, when I'm trying take-off in Loiter mode, lost PITCH and YAW control via transmitter.
I understand because of the required EKF GPS TYPE setting, Optical Flow in Arducopter excludes GPS functionality for the moment. But in Qgroudncontrol (or how would you call it) it is possible to get fused composition of GPS and Optical Flow data as it determines the most exact position from both.
Will there be anything like it added? You might even know when :-)?
the required GPS type setting when using Optical Flow prevents the use of a GPS system and waypoint navigation.
This is only the case for arducopter, not for qgroundcontrol. QGSC's firmware will fuse the GPS and optical flow data and calculate the most accurate position I understand.
Do you think GPS will be availalbe in arducopter with Optical flow anytime soon?
I have the issue whereby the PX4Flow does not work unless my PixHawk is powered by USB first. After spending some time going through various forums for answers, I found an example script to be placed in extras.txt to be executed at boot time.
I created the file /fs/microsd/etc/extras.txt with the following content:
if [ -f /bin/px4flow ] then sleep 4 if px4flow start then echo "Found px4flow sensor" fi fi
If I do a normal power up, px4flow is still not starting, but if I go in with nsh and run my script manually: sh /fs/microsd/etc/extras.txt, then px4flow starts with: [PX4FLOW] on I2C bus 1 at 0x42 (bus: 100KHz, max: 400KHz)and I get the message from the end of the script: Found px4flow sensor.
I'm quite familiar with UNIX type operating systems, but am just finding my way around NuttX. It looks to me like extras.txt might not be being executed at startup; Is this likely and what might be some reasons for it not executing? - I could be barking up the wrong tree, but I'm at a bit of a loss.
For anyone trying to fly indoors, I recommend laying down that foam padding that's used under carpeting, it has tons of surface texture (since it's the padding thats made of small chunks of multicolored foam) and it also provides some cushion for crashes.
Replies
Hi Paul,
This is an impressively stable test. Did you use the old firmware (px4flow-klt-06Dec2014) for this test flight or did you use a newer version from the PX4 repo? Cheers!
Hi everyone!
I'm testing v3.3.3 using PIXHAWK and PX4FLOW.
I read reply Goro Senzai and all comments. And quadrotor work is realy cool! But I still have question, when trying fly with coaxial octorotor. Octorotor, when I'm trying take-off in Loiter mode, lost PITCH and YAW control via transmitter.
Octorotor log in this reply. And quadro in next
Octorotor.zip
Quadro log
Quadro.zip
Randy's post lists the things not supported by the Navio+ and one of them is the Optical Flow.
Unfortunately, he doesn't give an explanation so does anyone know why it is not supported?
Hello Paul,
great work.
Will we see GPS fusion anytime soon?
I understand because of the required EKF GPS TYPE setting, Optical Flow in Arducopter excludes GPS functionality for the moment. But in Qgroudncontrol (or how would you call it) it is possible to get fused composition of GPS and Optical Flow data as it determines the most exact position from both.
Will there be anything like it added? You might even know when :-)?
Hello,
the required GPS type setting when using Optical Flow prevents the use of a GPS system and waypoint navigation.
This is only the case for arducopter, not for qgroundcontrol. QGSC's firmware will fuse the GPS and optical flow data and calculate the most accurate position I understand.
Do you think GPS will be availalbe in arducopter with Optical flow anytime soon?
sory because i am a begginer can you tell how can i run the script?
thanks
I have the issue whereby the PX4Flow does not work unless my PixHawk is powered by USB first. After spending some time going through various forums for answers, I found an example script to be placed in extras.txt to be executed at boot time.
I created the file /fs/microsd/etc/extras.txt with the following content:
if [ -f /bin/px4flow ]
then
sleep 4
if px4flow start
then
echo "Found px4flow sensor"
fi
fi
If I do a normal power up, px4flow is still not starting, but if I go in with nsh and run my script manually: sh /fs/microsd/etc/extras.txt, then px4flow starts with: [PX4FLOW] on I2C bus 1 at 0x42 (bus: 100KHz, max: 400KHz)and I get the message from the end of the script: Found px4flow sensor.
I'm quite familiar with UNIX type operating systems, but am just finding my way around NuttX. It looks to me like extras.txt might not be being executed at startup; Is this likely and what might be some reasons for it not executing? - I could be barking up the wrong tree, but I'm at a bit of a loss.
hi
i want a help please
i have bought a px4flow with ultrasonic and i am trying to connect with pixhawk
i update the firmware with qgroundcontrol and i set the focus
i set in misson planner
FLOW_ENABLE 1
FLOW_FXSCALER 0
FLOW_FYSCALER 0
FLOW_ORIENT_YAW -9000
EKF_GPS_TYPE 3
LOG_BITMASK = 131071
then i download the log but it dosent work
the OF.bodyX seems to follow the IMU.GyrX but the OF.flowX not
do you know what is the problem?
sorry for my english
thanks
printscreen.jpg
2016-02-08 15-28-41 110.bin.log.jpg