06-26-2023, 12:52 PM
Sorry, if this may seem a bit out of context, but Praveen contacted me with via email about your usecase and also pointed me to this thread. So, I am mostly here to post a QR code with multiple options. But let me quickly point out that the original inclination experiment only keeps one original value for each axis and the raw timestamp from the sensor. It then appends these time stamps to the actual list of values that are to be displayed when it calculates the angle. It does so to avoid unnecessary recalculations of the entire measurement, but this can also make things complicated if you change the data rate (because now it is limited by the analysis rate) and if you add another sensor, which is not synchronized by default and would have its own timestamps.
Now, since I also suggested to Praveen to also try the "attitude" sensor (a virtual sensor provided by the manufacturer of the phone, combining accelerometer, gyroscope and often the magnetometer to get an absolute angle, see https://phyphox.org/wiki/index.php/Attitude_sensor), I have attached a configuration with attitude, the inclination method from phyphox (just the atan of the accelerometer axes), the accelerometer and the gyroscope.
Note that the Euler angles use the common names used in aviation, but with the heading towards x and z being up. On a phone this is very unintuitive, because the equivalent to the inclination is now "yaw". But you will probably figure out the axes and if you are done testing and need a simplified and properly named version for students, I can simply provide one.
Also note that I made the average period for the inclination adjustable, because this really depends on your use case and if you also need the gyroscope I assume that the original setting would be too slow for you (again, something we can fixate once you know what you need/prefer). Unfortunately, it is not easy to provide accelerometer and gyroscope at maximum rate while providing averaging, so this might be a bit inefficient. If you are having trouble with phyphox or the measurement becoming unresponsive (especially if you lower the average period), let me know. In that case we should work with a fixed rate instead.
Now, since I also suggested to Praveen to also try the "attitude" sensor (a virtual sensor provided by the manufacturer of the phone, combining accelerometer, gyroscope and often the magnetometer to get an absolute angle, see https://phyphox.org/wiki/index.php/Attitude_sensor), I have attached a configuration with attitude, the inclination method from phyphox (just the atan of the accelerometer axes), the accelerometer and the gyroscope.
Note that the Euler angles use the common names used in aviation, but with the heading towards x and z being up. On a phone this is very unintuitive, because the equivalent to the inclination is now "yaw". But you will probably figure out the axes and if you are done testing and need a simplified and properly named version for students, I can simply provide one.
Also note that I made the average period for the inclination adjustable, because this really depends on your use case and if you also need the gyroscope I assume that the original setting would be too slow for you (again, something we can fixate once you know what you need/prefer). Unfortunately, it is not easy to provide accelerometer and gyroscope at maximum rate while providing averaging, so this might be a bit inefficient. If you are having trouble with phyphox or the measurement becoming unresponsive (especially if you lower the average period), let me know. In that case we should work with a fixed rate instead.