Determine the source of noise during circular-motion experiments

Determine whether the noise observed in the accelerometer signal during circular-motion experiments is caused by unintentional human hand movements, an imperfect rolling process, or another source, in order to distinguish genuine microadjustments from measurement noise suppressed by the Kalman filter.

Background

Experiment 5 evaluates the MPU6050-based tilt-estimation system during circular periodic motion. The accelerometer output approximately follows a sinusoidal pattern but contains spikes, while the Kalman filter smooths the signal into an almost perfect sinusoidal curve. The authors explicitly state that the source of the observed noise is unknown and identify two possible causes: involuntary movement by the person operating the apparatus or imperfections in the manual rolling process. Resolving this uncertainty is important because the filter may suppress intentional microadjustments as if they were measurement noise.

References

However, the source of noise is currently unknown; the noise could also be caused by the human's unintentional movement of the hand or an imperfect rolling process.

Design and Implementation of a Kalman Filter-Infused Algorithm for Tilt Estimation  (2609.00730 - Ma et al., 1 Sep 2026) in Section 6.5, Experiment 5 Result Analysis