Angle Estimation Using a Single Gyroscope with Self-Compensating Algorithm

Abstract

Conventional methods for estimating angles often depend on fuses of multiple sensors including gyroscopes, accelerometers, and magnetometers. In this article, a new method that uses only a single gyroscope to estimate the motion angles is presented. Although using multiple types of sensors is useful in drift correction, it can induce other issues such as reduced accuracy during rapid motion, prone to surrounding magnetic interference, increased computational burden, and hardly be incorporated into a low-cost microcontroller. The proposed method is based on a self-compensating algorithm that corrects the drifts using the gyroscope readings in real-time without the need for precalibration. Testing data is collected from 10 volunteers performing various physiotherapy exercises, and the proposed method results in an average correlation coefficient of 99.2% when compared to a marker-based optical motion capture system. The proposed algorithm is also computationally efficient in that it can be and has been incorporated into a low-cost microcontroller.