Sensor Fusion using Madgwick/Mahony/kalman filters the MATLAB coding

Hi all,

I have 6-DOF raw imu sensor data(only accelerometer and gyroscope). Now , wanted to fuse this data inorder to calculate 'Quaternions' and know the orientation. I am stuck at this point how to build a working algorithm in MATLAB of any of the above mentioned filters.

Any leads,references and already existing matlab scripts?? will be grateful.

Looking Forward.

Thanks.

Answer 1

[Amey Waghmare]

Hi,

As per my understanding, you have raw accelerometer and gyroscope data and want to obtain the Quaternion orientation estimates using a Sensor Fusion algorithm.

The Sensor Fusion and Tracking Toolbox contains ‘imufilter’ and ‘complementaryFilter’ objects to fuse accelerometer and magnetometer data. The ‘imufilter’ uses an internal error-state Kalman filter and the ‘complementaryFilter’ uses a complementary filter.

More details about the sensor fusion objects are available at the documentation;

• Imufilter: https://in.mathworks.com/help/fusion/ref/imufilter-system-object.html
• complementaryFilter: https://in.mathworks.com/help/fusion/ref/complementaryfilter-system-object.html

You can also refer to the following documentation to align and preprocess the raw sensor data: https://in.mathworks.com/help/fusion/ug/logged-sensor-data-alignment-for-orientation-estimation.html

Hope this resolves the issue.

Answer 2

[Remo Pillat]

Besides the filters that Amey mentioned, you can also find a more flexible inertial sensor fusion framework in the insEKF feature in Navigation Toolbox. This comes in handy if you want full control over the number and types of sensors to fuse, or if you want to use custom motion models and sensor models.

The following page provides an overview of how you can pick between the different fusion filters: https://www.mathworks.com/help/nav/ug/introduction-on-choosing-inertial-sensor-fusion-filters.html