the standard deviation of IMU sensor?

Discussions and questions about the Crazyflie Nano Quadcopter
Post Reply
DarkKnight
Beginner
Posts: 16
Joined: Tue Jun 15, 2021 10:19 pm

the standard deviation of IMU sensor?

Post by DarkKnight »

Hello, if anyone can help me the following questions, it would be very helpful for me:

Q1: Is the IMU frame aligned with the body frame of the quadcopter?

Q2:What is the standard deviation of accelerometer and gyroscope?

Q3: I have noticed that in your official website, the positive direction of roll and yaw angle are clockwise and the positive direction of pitch angle is counter-clockwise. However, I have done some tests. I hold the crazyflie and move it(Firstly I let crazyflie clockwise rotates along x axis 90 deg and back to 0, then counter-clockwise 90 deg along x axis and back to 0, then I let crazyflie clockwise rotates along y axis 90 deg and back to 0, then counter-clockwise 90 deg along y axis and back to 0, finally I let crazyflie clockwise rotates along z axis 90 deg and back to 0, then counter-clockwise 90 deg along z axis and back to 0). The results is attached in following picture(The estimated orientation is from kalman filter of the crazyflie). It shows that the although the positive direction of pitch angle is counter-clockwise, the positive direction of gyro in the y axis is clockwise, is that true?
Attachments
1.png
kimberly
Bitcraze
Posts: 852
Joined: Fri Jul 06, 2018 11:13 am

Re: the standard deviation of IMU sensor?

Post by kimberly »

Hi!

Q1. No I don't think that the IMU is not aligned but it's drivers make sure that the raw data is supposed aligned with the estimation.

Q2. The gyro std can be found here in the code. But the kalman filter does not need STD from the accelerometers. For precise measurements, I would refer to the product page of the component for documentatio but these are not measured in flight.

Q3. Yes exactly that is correct. You see that here in the code too: Why this is exactly done this way, I wouldn't know myself but I guess it seemed more logical at the time to have positive pitch be pointed up, even though we are using a east north up (ENU) coordinate system, as in the past they didn't want to go for the North east down (NED). But yes from an aerospace convention point of view it is a bit strange.
Post Reply