Optimization of adaptive I 2 H∞ control method based on multiple input sensors

With the growing demand for sustainable urban transportation, electric bicycles (E-bikes) have become an important solution due to their environmental friendliness and convenience. However, traditional control methods (such as PID and fuzzy logic) have limitations in dynamically adapting to rider behavior, environmental changes, and mechanical uncertainties, and it is difficult to balance accuracy, robustness, and rider comfort. To this end, this paper proposes a human-centric adaptive control framework that combines multi-sensor fusion with advanced control theory to achieve efficient coordination between rider intention and motor assistance. The core contributions of this study include: (1) Designing a multi-sensor current reference estimator to dynamically generate the optimal electromagnetic torque through state variables such as wheel speed, acceleration, slope, and human factor database (heart rate, subjective score, fatigue index) to achieve real-time prediction of rider demand; (2) Proposing an adaptive current reference value estimation algorithm that integrates feedforward compensation and error feedback to ensure smooth switching of assistance modes and suppress sensor noise; (3) Developing an intention-induced H∞ robust current tracking controller that significantly enhances the system’s robustness to parameter fluctuations and external disturbances by optimizing the H∞ norm of the closed-loop transfer function, while supporting personalized riding assistance.