Make a more powerful but still tiny drone.

Discussions about all things Bitcraze
olif
Beginner
Posts: 12
Joined: Wed Mar 18, 2020 5:21 am

Re: Make a more powerful but still tiny drone.

Post by olif » Thu Apr 30, 2020 7:14 pm

Thank you Tobias your hint was perfect! After correcting the vzkp and vzki values everything is working fine now. A little fine tuning of PID values is still open and having a look at the landing behaviour. How can I post a Video?
Attachments
Bolt airborne.jpg
crazyflie Bolt is airborne



tobias
Bitcraze
Posts: 1992
Joined: Mon Jan 28, 2013 7:17 pm
Location: Sweden

Re: Make a more powerful but still tiny drone.

Post by tobias » Mon May 04, 2020 1:21 pm

Great news! Looks like a cool drone ready to do research on. Any problems with the laser ranger sensors, e.g. detecting the props or similar?

If you want to post a video you will have to link it from e.g. youtube. Our server is to weak to have it...sorry

olif
Beginner
Posts: 12
Joined: Wed Mar 18, 2020 5:21 am

Re: Make a more powerful but still tiny drone.

Post by olif » Fri May 22, 2020 6:33 am

I dont have issues with the ToF sensors. They are not influenced by the propellers. My next topic is how can the bolt fly inside a room in about 1,5m height starting from ground and then crossing tables and other obstacles without jumping up because of height measurement from the flow deck. How can I disable the height measurement from the flow deck and enable the baro for height measurement? Is the behavior regarding hight holding now better with the bosch sensor? Have you done further tests in this direction? If there is no solution, this will be a killing point. I can not use the top sensor for hight measurement because the ceiling is cluttered.

kimberly
Bitcraze
Posts: 323
Joined: Fri Jul 06, 2018 11:13 am

Re: Make a more powerful but still tiny drone.

Post by kimberly » Fri May 22, 2020 7:36 am

Currently there is an github issue about this feature: https://github.com/bitcraze/crazyflie-f ... issues/540.

It is not implemented into the firmware yet but it might give some pointers on how to fix this. This is indeed a common problem in a lot of opensource autopilot programs so there is not an proven fix for this, so it would be good to rekindle the discussion there.

Another thing that might be added, is that you do the height control in velocity mode only.

Post Reply