Development of Automatic Gear Shifting for Bicycle Riding Based on Physiological Information and Environment Sensing

This paper presents the development of an intelligent bicycle system with automatic gear shifting based on the user’s physiological information and environment sensing. The sensor data are collected during the ride, uploaded to the cloud database, and analyzed using random forest to derive the appropriate riding requirement. Our proposed technique performs the automatic gear shifting operation using inexpensive sensor components and embedded system development boards. The gear shifting module is designed and installed as an external device for the off-the-shelf bicycle, without the need to dismantle or modify the body structure. A bicycle riding specific online management interface is developed for web access and administration. It allows the cloud data being displayed and let users acquire and understand their riding behaviors. The experiments and evaluation in the outdoor environment have demonstrated the feasibility of the proposed intelligent gear shifting system.