Estimation of pedaling torque for electric power-assisted bicycle on slope environment

In recent years, electric power assisted bicycles become popular in Japan. However the prices of such bicycles are still high because an expensive torque sensor is attached. It is a necessary component to measure human pedaling torque for comfortable motor assistance. Therefore realization of pedaling torque estimation without torque sensor is effective for the cost reduction. In our previous studies, we did not consider the condition for travelling on upward slope. In order to increase accuracy of the estimation regardless of road environmental change, this paper presents an improved method for pedaling torque estimation using the Recursive Least Square (RLS) method with multiple forgetting factors. The validity of the proposed method is verified by simulations and experiments.