Development of an Adaptive Speed Control System for Pedal-Assisted Electric Bikes

This paper presents the development and experimental validation of an adaptive speed control system for pedal-assist electric bicycles (e-bikes). Conventional e-bikes, which rely on fixed assistance levels, often provide a suboptimal riding experience. To address this, we designed a system centered on a Fuzzy-Proportional-Integral (Fuzzy-PI) controller. This system leverages real-time data from a torque sensor and a speed sensor to dynamically adjust the motor's power output, aiming to maintain a rider-set target speed seamlessly. The prototype e-bike, built around a 500W motor and an STM32F4 microcontroller, was tested against a conventional PI controller. Results demonstrate that the adaptive Fuzzy-PI system offers superior performance, including faster response to load changes (~0.1s vs. ~0.3s settling time), minimal speed overshoot (<2% vs. >7%), and improved energy efficiency. This research confirms that intelligent, adaptive control can significantly enhance rider comfort, safety, and battery life in personal mobility solutions.