Estimation of Torque Exerted by Cyclists on Electric Bicycles with Pedal Assist Using Disturbance Observer

In this work, a strategy is proposed for estimating the torque exerted by a cyclist on an electric bicycle using an integral disturbance observer. The approach is based on the longitudinal dynamic model of the bicycle, which includes the resistive load. Using this model, an estimation strategy is proposed that utilizes the available measurements from the permanent magnet synchronous motor (PMSM), where the angular velocity is a state variable of the model and is determined from the Hall effect sensors located in the motor's stator. This estimation strategy presents a novel approach, as it can be applied to a pedal-assist bicycle without the need for a torque sensor, significantly reducing costs while simultaneously increasing the robustness and reliability of the system. Finally, to validate the proposed observer, experimental results are presented using a 26-inch electric bicycle ridden by two riders of different weights, with a torque sensor integrated into the pedal axle.