Online power meter calibration for accurate cyclist power control

This paper deals with cyclist power estimation in bicycles via a power meter device measuring the torque applied to the pedal. In order to guarantee the correctness of power estimation, it can be shown that the angular displacement between the crank arm longitudinal axis and the torque measurement axis must be accurately known. In this work, an online algorithm is developed to estimate such an angle, which can also vary along the time due to mechanical stress, without the need of a periodic calibration procedure. All these features make the resulting power estimation suitable for real-time cyclist power control. Experimental results show the effectiveness of the proposed estimation procedure on real road tests.