SensorFusion
This project performs sensor fusion to track a mobile device's orientation. The data utilised are from three sensors: a) Accelerometer, b) Magnetic Field, d) Gyroscope. The sensor fusion is executed off-line. This project basically ports code developed by Paul Lawitzki from Android to Matlab/Octave. The Matlab/Octave code imports a CSV file with a given structure. Then, a strapdown integration system is developed by computing the orientation from two different components: a) Accelerometer - Magnetic Field and b) Gravity tracking through Gyroscope. Finally, the orientations from the two components are fused.
Install / Use
/learn @palaghias/SensorFusionREADME
SensorFusion
This project performs sensor fusion to track a mobile device's orientation. The data utilised are from three sensors: a) Accelerometer, b) Magnetic Field, c) Gyroscope. The sensor fusion is executed off-line.
This project basically ports code developed by Paul Lawitzki from Android to Matlab/Octave.The Matlab/Octave code imports a CSV file with a given structure. Then, a strapdown integration system is developed by computing the orientation from two different components:
a) Accelerometer - Magnetic Field and b) Gravity tracking through Gyroscope.
Finally, the orientations from the two components are fused. (http://www.thousand-thoughts.com/2012/03/android-sensor-fusion-tutorial/)
The developer should note that testSensorFusion imports the data from a CSV file that follows a particular structure; the import method should be modified otherwise.
How-to
-
Clone the project in a directory and navigate Matlab/Octave to the particular directory.
-
Run the test file: source-code/sensor_fusion/testSensorFusion.m
