SMART E-BICYCLE

Continuous performance monitoring and rider safety are critical for the effective adoption of electric micro-mobility systems in urban environments. This paper presents a Smart E-Bicycle framework designed for real-time monitoring and analysis using an embedded platform. The system integrates a sensor front-end, data acquisition unit, signal preprocessing, feature extraction, and a lightweight rule-based decision engine. The proposed architecture continuously captures key electrical and mechanical parameters such as battery voltage, current, speed, and vibration. A signal-quality-aware preprocessing stage enhances data reliability by reducing noise and improving measurement accuracy. A hybrid processing pipeline is then used to estimate system states and extract meaningful features representing the bicycle’s operational condition. These features are evaluated using a rule-based decision mechanism to detect faults and identify unsafe riding conditions in real time. The system is optimized for low computational complexity and power efficiency, making it suitable for embedded deployment. Experimental results demonstrate reliable abnormality detection with high accuracy and minimal processing overhead. The proposed framework improves rider safety, enhances system reliability, and supports the advancement of intelligent urban mobility solutions.