Iris: Highly Erratic Behaviour Midflight

Hello - 

I took my Iris out today for a survey and had a very scary experience. I've had the Iris for almost half a year now and have flown many missions and have logged many hours of manual flying, so I'm a fairly competent flier. I normally fly missions, but since I was surveying a small area, I decided to fly the Iris manually (in loiter mode) A couple of seconds after take off, the Iris became very hard to control and starting listing strongly towards one direction. I had to yank the control stick all the way towards the opposite direction to keep it from flying away. I desperately tried to land the Iris, but it was hard to do so as it was flying very erratically. I switched to alt hold mode after a few seconds to see if that help. Thankfully the Iris became a bit better behaved, but still almost impossible to control. After many nervous seconds, I was finally able to land the Iris into a bush. I'm so glad I didn't fly into a power line or hit a house. Some other observations:

- The yaw and altitude control were working ok

- The roll and pitch was not. The quad strongly wanted to fly in one direction

- The behaviour became slightly better after switching to altitude hold mode, though still almost impossible to control. 

Thankfully I have the flight log (attached). You can see the erratic flight path..

I also had a GoPro attached and it was taking time lapse pics. Happy to send those as well if it helps with debugging. 

Really thankful that I didn't run into something and was able to recover the Iris with only minor damage. I really would like to understand what happened here. Malfunctioning orientation sensor/GPS? I ran the log through mission planner and except for a DCM Bad Heading message towards the end of the log, don't see any other error messages.

I'd really appreciate if someone can help me diagnose what's happening here. I don't want to fly again until I know what happened with this flight. 

Thanks,

Ankur

2015-08-09 19-51-10.tlog

2015-08-09 19-51-10.rlog

You need to be a member of diydrones to add comments!

Join diydrones

Email me when people reply –

Replies

  • Ankur,

    I had a similar experience.  I had just done a flight for several minutes, flying manually in Loiter mode like you.  Everything was going great.  I brought it back and turned on my GoPro to collect some test video.  On my second flight, at about 80 ft I had it Loitering and was rotating to get 360 degree view of video when it suddenly veered to one side.  I was able to stabilize and bring it close to the ground under decent control.  Then I tried to rotate again and it veered strongly and crashed, bending a prop. 

    If anyone has an explanation, I would love to hear what is causing this issue and whether it's sensor based or something else.  This poses a safety concern if it's happening regularly.

  • Bump. Anyone? I'd think 3DR engineers would be interested in helping to diagnose erratic flight logs? 

  • Hello: Update - I partially got over some of the jitters from the scary experience yesterday and did another flight test today in a big park. I flew the quad very close to the ground so I could land it quickly in case anything went wrong - tried loiter and alt hold and then did a small mission. Flight log attached. It all went smoothly, so there doesn't seem anything wrong with the hardware. I'm very curious to understand what went so wrong yesterday. I tried comparing the flight log from yesterday with the one today and looked at various parameters, particularly the mag/gyro/accel data - see attached word doc for a screenshot of the plots of accx, accy, accz. Also attached is the flight log from today. I'd appreciate it if someone can help point me to where I should investigate to identify what could potentially be wrong. 

    Good_Log.tlog

    Good_Bad_Flight_Comparison.docx

    https://storage.ning.com/topology/rest/1.0/file/get/3702067052?profile=original
This reply was deleted.

Activity