[SOLVED] Crazyflie 2.0 stability issue or working as intended?

[mhinojosa]

Hello Everyone,

I recently bought the Crazyflie 2.0 for myself for Christmas and have been playing around with some of the Python examples to get a better idea of how the flie performs before I move on to some other projects with it.

I made some changes to the ramp.py example to just get the flie off the ground to see how it handles, i merely increased the thrust values and kept all other variables at 0. My first attempt had the flie pitch backwards and roll right as it took off. I was able to fix this by changing the values in the client config tool and now the flie takes off straight enough into the air for my purposes but it does not stay around a point in space. I know the flie cannot hover on its own but I was thinking that it can stay somewhat in position while in the air with no extra commands given to it other then thrust values. I have tried playing around with the config values but can't seem to get the flie to not drift off in some direction. Is this normal behavior for the flie or might I have some other issue, my console log output looks normal and the propellers are balanced as far as I can tell.

I hope this is problemy enough for the support forum as I am looking to buy another flie but I want to make sure the one I have now is behaving as intended or if there might be some other problem.

Thank you for any help you might be able to give me.

[chad]

My first attempt had the flie pitch backwards and roll right as it took off. I was able to fix this by changing the values in the client config tool and now the flie takes off straight enough into the air for my purposes but it does not stay around a point in space.

There is rotor wash near the ground level that disturbs the air considerably. This can cause attitude changes at take-off. Also, battery placement can impact the balance of Crazyflie. The need to set trim values for pitch and roll is not uncommon.

I was thinking that it can stay somewhat in position while in the air with no extra commands given to it other then thrust values. I have tried playing around with the config values but can't seem to get the flie to not drift off in some direction. Is this normal behavior for the flie or might I have some other issue, my console log output looks normal and the propellers are balanced as far as I can tell.

This is fairly normal behavior. Quads of this size are more susceptible to minor changes in air current and pressure and without GPS a position hold is really tough. You'll not be able to precisely control the position of the Crazyflie with thrust alone, it will take pitch, roll, and yaw input is well. Yaw will also slowly drift, so over time you will also need to compensate for that. Bitcraze has been working on a local positioning system that could make what you're trying to do (completely autonomous flight) more easily obtainable. Others have created camera based solutions (and here) as well. You can still do a fair amount of autonomy without these larger control systems though. Infrared distance sensors, ultrasound, and other sensors have been used for some application successfully.

[mhinojosa]

Thank you very much chad, this information has been really helpful to me, knowing that is fairly typical behavior has definitely put me at ease on the build and what the flie is capable of. Just picked up my second and third crazyflies so I am excited to get started on some kind of positioning system for my flies. Thanks again for the reply.

[chad]

Glad I could provide some meaningful info! Post back on the progress of what you're doing, I'm sure others will be interested.