IMU for Model Rocket (2000ft apogee) : 6dof IMU or 9dof IMU?

[niranjandr]

Greetings to the community,

We were working on a model rocket at the university and we wanted to collect IMU data from the first flight, to help design a Kalman filter for sensor fusion of altimeter and IMU in future missions.

I wanted to know whether the accelerometer and gyro data from a 6dof IMU is enough for apogee detection in a model rocket or is the additional magnetometer data required (9dof IMU)? Will the use of a 9dof IMU increase the accuracy in any way?

Thanks in advance

 

[Voyager1]

10 DoF BMO055 + BMP280 Gravity IMU is a good choice. It provides integrated data for quaternions, Euler angles or vectors, etc. Additionally, you have a barometric sensor too.

https://wiki.dfrobot.com/Gravity_BNO055_+_BMP280 intelligent_10DOF_AHRS_SKU_SEN0253

 

[UhClem]

The magnetometer data is only useful if you expect a significant shift in rate gyro zero offsets during the flight. Which is unlikely as the flight is too short.

In any case an EKF is overkill for apogee detection. (I assume you will be using an EKF.)

 

[OverTheTop]

The magnetometer may also be useful if your rocket rolls at a rate quicker than the gyro can measure. Not likely a problem if you build the rocket straight .

6dof is adequate, but the data from a magnetometer can be interesting too. I would include it IMHO.

 

[Ravenex]

10 DoF BMO055 + BMP280 Gravity IMU is a good choice. It provides integrated data for quaternions, Euler angles or vectors, etc. Additionally, you have a barometric sensor too.

https://wiki.dfrobot.com/Gravity_BNO055_+_BMP280 intelligent_10DOF_AHRS_SKU_SEN0253

This seems to have a maximum acceleration of +/-16Gs which seems like it would be too low for most rocket applications.

 

[OverTheTop]

This seems to have a maximum acceleration of +/-16Gs which seems like it would be too low for most rocket applications.

Quite a few designs use one of these and also a higher 200G version to cover the required range. That way you can get the nicer resolution of the 16G unit and the overall range of the 200G.

You could also use a 200G unit and average the readings, if you can read fast enough, to increase the resolution and improve the noise. You have to be able to process it quick enough though.

 

[Voyager1]

This seems to have a maximum acceleration of +/-16Gs which seems like it would be too low for most rocket applications.

Yes, I agree that the +/-16 G accelerometer is a limitation for most mid-high power rockets, but as OTT correctly points out, some designs (e.g., Altus Metrum, Aim Xtra, etc) do employ 16 G accelerometers for greater resolution and have extra primary axis 100-200 G sensors to get the higher range. The OP did specify a 2,000' apogee, which doesn't indicate a particularly high G flight, but point taken.

However, the application that I have been using this IMU in is for relatively low-powered (i.e., low & slow) water rocket test flights for high school science mentoring activities. The wealth of information from this one little IMU makes it ideal for these applications.

 

[niranjandr]

The magnetometer data is only useful if you expect a significant shift in rate gyro zero offsets during the flight. Which is unlikely as the flight is too short.

In any case an EKF is overkill for apogee detection. (I assume you will be using an EKF.)

Thanks a lot, for the insight. Yeah we were planning to use a normal Kalman filter, as EKF would be overkill for our purpose, as you mentioned

 

[jderimig]

1dof is all you need for apogee detection.