Guaranteed stability fuzzy controller for stabilization the motion of an unmanned bicycle is proposed. First, a fuzzy control system capable of automatically balancing an unmanned bicycle through tracking desired roll angle is developed. Fuzzy logic controller membership functions are defined utilizing scaling factors. To guarantee the stability of the closed loop system, similar to previous approaches reported in the literature, fuzzy If–Then rules are constructed based on Lyapunov stability criterion. It is indicated that the proposed fuzzy controller violates Lyapunov stability criterion. The reason of such a violation is argued in detail. To cope with this shortcoming, some modifications are made to the control strategy to assure stability. Through these modifications, the modified fuzzy controller is developed which simultaneously balances the bicycle and guarantees stability while minimizing roll angle tracking error and its derivative. It is indicated that the improved fuzzy controller can adapt to a variety of initial conditions. Moreover, robustness of the controller against parameter variation is verified through its implementation on different bicycle designs (different sets of bicycle parameters). Simulation results confirm the efficacy of the proposed fuzzy controller in terms of settling time and overshoot in comparison with previous studies. Sensitivity analysis of the controller efficiency with respect to system parameters is also assessed.